id
string | submitter
string | authors
string | title
string | comments
string | journal-ref
string | doi
string | report-no
string | categories
string | license
string | abstract
string | versions
list | update_date
timestamp[s] | authors_parsed
sequence |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
2504.19594 | Lorenzo Alvisi | Lorenzo Alvisi, Serena Tardelli, Maurizio Tesconi | Mapping the Italian Telegram Ecosystem: Communities, Toxicity, and Hate
Speech | null | null | null | null | cs.SI cs.AI | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Telegram has become a major space for political discourse and alternative
media. However, its lack of moderation allows misinformation, extremism, and
toxicity to spread. While prior research focused on these particular phenomena
or topics, these have mostly been examined separately, and a broader
understanding of the Telegram ecosystem is still missing. In this work, we fill
this gap by conducting a large-scale analysis of the Italian Telegram sphere,
leveraging a dataset of 186 million messages from 13,151 chats collected in
2023. Using network analysis, Large Language Models, and toxicity detection
tools, we examine how different thematic communities form, align ideologically,
and engage in harmful discourse within the Italian cultural context. Results
show strong thematic and ideological homophily. We also identify mixed
ideological communities where far-left and far-right rhetoric coexist on
particular geopolitical issues. Beyond political analysis, we find that
toxicity, rather than being isolated in a few extreme chats, appears widely
normalized within highly toxic communities. Moreover, we find that Italian
discourse primarily targets Black people, Jews, and gay individuals
independently of the topic. Finally, we uncover common trend of intra-national
hostility, where Italians often attack other Italians, reflecting regional and
intra-regional cultural conflicts that can be traced back to old historical
divisions. This study provides the first large-scale mapping of the Italian
Telegram ecosystem, offering insights into ideological interactions, toxicity,
and identity-targets of hate and contributing to research on online toxicity
across different cultural and linguistic contexts on Telegram.
| [
{
"version": "v1",
"created": "Mon, 28 Apr 2025 08:58:18 GMT"
},
{
"version": "v2",
"created": "Mon, 5 May 2025 09:35:11 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Alvisi",
"Lorenzo",
""
],
[
"Tardelli",
"Serena",
""
],
[
"Tesconi",
"Maurizio",
""
]
] |
2504.19718 | Daoye Wang | Victoria Yue Chen, Daoye Wang, Stephan Garbin, Jan Bednarik, Sebastian
Winberg, Timo Bolkart, Thabo Beeler | Pixels2Points: Fusing 2D and 3D Features for Facial Skin Segmentation | 4 pages, 4 figures, to be published in Eurographics 2025 as a short
paper | null | null | null | cs.GR cs.CV | http://creativecommons.org/licenses/by/4.0/ | Face registration deforms a template mesh to closely fit a 3D face scan, the
quality of which commonly degrades in non-skin regions (e.g., hair, beard,
accessories), because the optimized template-to-scan distance pulls the
template mesh towards the noisy scan surface. Improving registration quality
requires a clean separation of skin and non-skin regions on the scan mesh.
Existing image-based (2D) or scan-based (3D) segmentation methods however
perform poorly. Image-based segmentation outputs multi-view inconsistent masks,
and they cannot account for scan inaccuracies or scan-image misalignment, while
scan-based methods suffer from lower spatial resolution compared to images. In
this work, we introduce a novel method that accurately separates skin from
non-skin geometry on 3D human head scans. For this, our method extracts
features from multi-view images using a frozen image foundation model and
aggregates these features in 3D. These lifted 2D features are then fused with
3D geometric features extracted from the scan mesh, to then predict a
segmentation mask directly on the scan mesh. We show that our segmentations
improve the registration accuracy over pure 2D or 3D segmentation methods by
8.89% and 14.3%, respectively. Although trained only on synthetic data, our
model generalizes well to real data.
| [
{
"version": "v1",
"created": "Mon, 28 Apr 2025 12:13:12 GMT"
},
{
"version": "v2",
"created": "Sun, 4 May 2025 10:27:59 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Chen",
"Victoria Yue",
""
],
[
"Wang",
"Daoye",
""
],
[
"Garbin",
"Stephan",
""
],
[
"Bednarik",
"Jan",
""
],
[
"Winberg",
"Sebastian",
""
],
[
"Bolkart",
"Timo",
""
],
[
"Beeler",
"Thabo",
""
]
] |
2504.19789 | Tianyun Wei | Tianyun Wei, Zechen Lan, Yasunobu Arikawa, Yanjun Gu, Takehito
Hayakawa, Alessio Morace, Ryuya Yamada, Kohei Yamanoi, Koichi Honda, Masaki
Kando, Nakanii Nobuhiko, Seyed Reza Mirfayzi, Sergei V. Bulanov and Akifumi
Yogo | Quasi-monoenergetic Deuteron Acceleration via Boosted Coulomb Explosion
by Reflected Picosecond Laser Pulse | null | null | null | null | physics.plasm-ph hep-ex | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Generation of quasi-monoenergetic ions by intense laser is one of
long-standing goals in laser-plasma physics. However, existing laser-driven ion
acceleration schemes often produce broad energy spectra and limited control
over ion species. Here we propose the acceleration mechanism, boosted Coulomb
explosion, initiated by a standing wave, which is formed in a pre-expanded
plasma by the interference between a continuously incoming main laser pulse and
the pulse reflected by a solid target, where the pre-expanded plasma is formed
from a thin layer on the solid target by a relatively strong pre-pulse. This
mechanism produces a persistent Coulomb field on the target front side with
field strengths on the order of TV/m for picoseconds. We experimentally
demonstrate generation of quasi-monoenergetic deuterons up to 50 MeV using an
in-situ D$_2$O-deposited target. Our results show that the peak energy can be
tuned by the laser pulse duration.
| [
{
"version": "v1",
"created": "Mon, 28 Apr 2025 13:33:10 GMT"
},
{
"version": "v2",
"created": "Mon, 5 May 2025 06:56:47 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Wei",
"Tianyun",
""
],
[
"Lan",
"Zechen",
""
],
[
"Arikawa",
"Yasunobu",
""
],
[
"Gu",
"Yanjun",
""
],
[
"Hayakawa",
"Takehito",
""
],
[
"Morace",
"Alessio",
""
],
[
"Yamada",
"Ryuya",
""
],
[
"Yamanoi",
"Kohei",
""
],
[
"Honda",
"Koichi",
""
],
[
"Kando",
"Masaki",
""
],
[
"Nobuhiko",
"Nakanii",
""
],
[
"Mirfayzi",
"Seyed Reza",
""
],
[
"Bulanov",
"Sergei V.",
""
],
[
"Yogo",
"Akifumi",
""
]
] |
2504.19807 | Mohit Sharma | Mohit K. Sharma, Leandros Perivolaropoulos, and M. Sami | Multi-Parameter Transitions in Cosmological Calibrators: A Resolution to
the Hubble Tension from SH0ES Data Analysis | 40 pages, 6 figures, minor revisions, updated references. Comments
are welcome | null | null | null | astro-ph.CO gr-qc | http://creativecommons.org/licenses/by/4.0/ | The Hubble tension, characterized by discrepant measurements of the Hubble
constant from early and late universe probes, remains one of the most
significant challenges in cosmology. Building upon our previous analysis of
individual parameter transitions in SH0ES data, we investigate the impact of
simultaneous transitions in multiple Cepheid and SNIa calibration parameters at
specific cosmic distances. We allow various combinations of transitions in
Cepheid absolute magnitude ($M^W_H$), period-luminosity relation slope ($b_W$),
metallicity coefficient ($Z_W$), and SNIa absolute magnitude ($M_B$). Our
comprehensive analysis reveals a consistent preferred transition distance of
approximately 23 Mpc across different parameter combinations. The most
statistically favored model allows simultaneous transitions in $b_W$, $Z_W$,
and $M_B$, yielding $\Delta \text{AIC} \simeq -9.2$ and $\Delta \text{BIC}
\simeq -3.0$ compared to the baseline SH0ES model. This provides strong
evidence for inhomogeneities in standard candle calibrations. We demonstrate
that the post-transition SNIa absolute magnitude aligns more closely with
CMB-based constraints, resulting in a reduced Hubble constant value that
alleviates the tension. Our findings suggest that the Hubble tension might be
resolved through proper modeling of calibration parameter inhomogeneities
rather than requiring new physics beyond $\Lambda$CDM.
| [
{
"version": "v1",
"created": "Mon, 28 Apr 2025 14:04:27 GMT"
},
{
"version": "v2",
"created": "Fri, 2 May 2025 18:13:19 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Sharma",
"Mohit K.",
""
],
[
"Perivolaropoulos",
"Leandros",
""
],
[
"Sami",
"M.",
""
]
] |
2504.19820 | YoonHyuk Choi | Yoonhyuk Choi, Jiho Choi, Taewook Ko, Chong-Kwon Kim | Hierarchical Uncertainty-Aware Graph Neural Network | null | null | null | null | cs.LG cs.IR | http://creativecommons.org/licenses/by/4.0/ | Recent research on graph neural networks (GNNs) has explored mechanisms for
capturing local uncertainty and exploiting graph hierarchies to mitigate data
sparsity and leverage structural properties. However, the synergistic
integration of these two approaches remains underexplored. This work introduces
a novel architecture, the Hierarchical Uncertainty-Aware Graph Neural Network
(HU-GNN), which unifies multi-scale representation learning, principled
uncertainty estimation, and self-supervised embedding diversity within a single
end-to-end framework. Specifically, HU-GNN adaptively forms node clusters and
estimates uncertainty at multiple structural scales from individual nodes to
higher levels. These uncertainty estimates guide a robust message-passing
mechanism and attention weighting, effectively mitigating noise and adversarial
perturbations while preserving predictive accuracy on semi-supervised
classification tasks. We also offer key theoretical contributions, including a
probabilistic formulation, rigorous uncertainty-calibration guarantees, and
formal robustness bounds. Extensive experiments on standard benchmarks
demonstrate that our model achieves state-of-the-art robustness and
interpretability.
| [
{
"version": "v1",
"created": "Mon, 28 Apr 2025 14:22:18 GMT"
},
{
"version": "v2",
"created": "Mon, 5 May 2025 07:47:54 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Choi",
"Yoonhyuk",
""
],
[
"Choi",
"Jiho",
""
],
[
"Ko",
"Taewook",
""
],
[
"Kim",
"Chong-Kwon",
""
]
] |
2504.19919 | Erbo Li | Erbo Li, Qi Qin, Yifan Sun, and Liping Zhu | Distributed Reconstruction from Compressive Measurements: Nonconvexity
and Heterogeneity | null | null | null | null | stat.ME | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The compressive sensing (CS) and 1-bit CS demonstrate superior efficiency in
signal acquisition and resource conservation, while 1-bit CS achieves maximum
resource efficiency through sign-only measurements. With the emergence of
massive data, the distributed signal aggregation under CS and 1-bit CS
measurements introduces many challenges, including nonconvexity and
heterogeneity. The nonconvexity originates from the unidentifiability of signal
magnitude under finite-precision measurements. The heterogeneity arises from
the signal and noisy measurement on each node. To address these challenges, we
propose a framework with a squared cosine similarity penalty. We address
nonconvexity by an novel invex relaxation formulation to ensure the uniqueness
of the global optimality. For heterogeneous signals and noisy measurements, the
proposed estimate adaptively debiases through correction guided by similarity
and signal-to-noise ratio (SNR) information. Our method achieves a high
probability minimax-optimal convergence rate under sufficient node counts and
similarity conditions, improving from $O\{(p\log{p}/n_j)^{1/2}\}$ to
$O\{(p\log{p}/N)^{1/2}+p^{1/2}/n_j\}$, with signal dimension $p$, local and
total sizes $n_j$ and $N$. Extensive simulations validate the method's
effectiveness and performance gains in reconstructing heterogeneous signals
from 1-bit CS measurements. The proposed framework maintains applicability to
CS measurements while reducing communication overhead in distributed setting.
| [
{
"version": "v1",
"created": "Mon, 28 Apr 2025 15:48:18 GMT"
},
{
"version": "v2",
"created": "Sun, 4 May 2025 08:05:05 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Li",
"Erbo",
""
],
[
"Qin",
"Qi",
""
],
[
"Sun",
"Yifan",
""
],
[
"Zhu",
"Liping",
""
]
] |
2504.19956 | Vineeth Sai Narajala | Vineeth Sai Narajala, Om Narayan | Securing Agentic AI: A Comprehensive Threat Model and Mitigation
Framework for Generative AI Agents | 12 pages, 2 figures, 1 table, typos corrected, references added | null | null | null | cs.CR cs.AI | http://creativecommons.org/licenses/by/4.0/ | As generative AI (GenAI) agents become more common in enterprise settings,
they introduce security challenges that differ significantly from those posed
by traditional systems. These agents are not just LLMs; they reason, remember,
and act, often with minimal human oversight. This paper introduces a
comprehensive threat model tailored specifically for GenAI agents, focusing on
how their autonomy, persistent memory access, complex reasoning, and tool
integration create novel risks. This research work identifies 9 primary threats
and organizes them across five key domains: cognitive architecture
vulnerabilities, temporal persistence threats, operational execution
vulnerabilities, trust boundary violations, and governance circumvention. These
threats are not just theoretical they bring practical challenges such as
delayed exploitability, cross-system propagation, cross system lateral
movement, and subtle goal misalignments that are hard to detect with existing
frameworks and standard approaches. To help address this, the research work
present two complementary frameworks: ATFAA - Advanced Threat Framework for
Autonomous AI Agents, which organizes agent-specific risks, and SHIELD, a
framework proposing practical mitigation strategies designed to reduce
enterprise exposure. While this work builds on existing work in LLM and AI
security, the focus is squarely on what makes agents different and why those
differences matter. Ultimately, this research argues that GenAI agents require
a new lens for security. If we fail to adapt our threat models and defenses to
account for their unique architecture and behavior, we risk turning a powerful
new tool into a serious enterprise liability.
| [
{
"version": "v1",
"created": "Mon, 28 Apr 2025 16:29:24 GMT"
},
{
"version": "v2",
"created": "Fri, 2 May 2025 18:42:42 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Narajala",
"Vineeth Sai",
""
],
[
"Narayan",
"Om",
""
]
] |
2504.20081 | Rahul Vishwakarma | Rahul Vishwakarma and Sinchan Banerjee | Billions at Stake: How Self-Citation Adjusted Metrics Can Transform
Equitable Research Funding | 8 Pages | null | null | null | cs.DL | http://creativecommons.org/licenses/by/4.0/ | Citation metrics serve as the cornerstone of scholarly impact evaluation
despite their well-documented vulnerability to inflation through self-citation
practices. This paper introduces the Self-Citation Adjusted Index (SCAI), a
sophisticated metric designed to recalibrate citation counts by accounting for
discipline-specific self-citation patterns. Through comprehensive analysis of
5,000 researcher profiles across diverse disciplines, we demonstrate that
excessive self-citation inflates traditional metrics by 10-20%, potentially
misdirecting billions in research funding. Recent studies confirm that
self-citation patterns exhibit significant gender disparities, with men
self-citing up to 70% more frequently than women, exacerbating existing
inequalities in academic recognition. Our open-source implementation provides
comprehensive tools for calculating SCAI and related metrics, offering a more
equitable assessment of research impact that reduces the gender citation gap by
approximately 8.5%. This work contributes to the paradigm shift toward
transparent, nuanced, and equitable research evaluation methodologies in
academia, with direct implications for funding allocation decisions that
collectively amount to over $100 billion annually in the United States alone.
| [
{
"version": "v1",
"created": "Fri, 25 Apr 2025 09:20:09 GMT"
},
{
"version": "v2",
"created": "Sun, 4 May 2025 02:44:28 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Vishwakarma",
"Rahul",
""
],
[
"Banerjee",
"Sinchan",
""
]
] |
2504.20110 | Yu-Hsuan Chen | Yu-hsuan Chen, Jing Bi, Cyril Ngo Ngoc, Victor Oancea, Jonathan Cagan,
Levent Burak Kara | Attention to Detail: Fine-Scale Feature Preservation-Oriented Geometric
Pre-training for AI-Driven Surrogate Modeling | null | null | null | null | cs.LG | http://creativecommons.org/licenses/by-nc-nd/4.0/ | AI-driven surrogate modeling has become an increasingly effective alternative
to physics-based simulations for 3D design, analysis, and manufacturing. These
models leverage data-driven methods to predict physical quantities
traditionally requiring computationally expensive simulations. However, the
scarcity of labeled CAD-to-simulation datasets has driven recent advancements
in self-supervised and foundation models, where geometric representation
learning is performed offline and later fine-tuned for specific downstream
tasks. While these approaches have shown promise, their effectiveness is
limited in applications requiring fine-scale geometric detail preservation.
This work introduces a self-supervised geometric representation learning method
designed to capture fine-scale geometric features from non-parametric 3D
models. Unlike traditional end-to-end surrogate models, this approach decouples
geometric feature extraction from downstream physics tasks, learning a latent
space embedding guided by geometric reconstruction losses. Key elements include
the essential use of near-zero level sampling and the innovative batch-adaptive
attention-weighted loss function, which enhance the encoding of intricate
design features. The proposed method is validated through case studies in
structural mechanics, demonstrating strong performance in capturing design
features and enabling accurate few-shot physics predictions. Comparisons with
traditional parametric surrogate modeling highlight its potential to bridge the
gap between geometric and physics-based representations, providing an effective
solution for surrogate modeling in data-scarce scenarios.
| [
{
"version": "v1",
"created": "Sun, 27 Apr 2025 17:10:13 GMT"
},
{
"version": "v2",
"created": "Sat, 3 May 2025 12:01:02 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Chen",
"Yu-hsuan",
""
],
[
"Bi",
"Jing",
""
],
[
"Ngoc",
"Cyril Ngo",
""
],
[
"Oancea",
"Victor",
""
],
[
"Cagan",
"Jonathan",
""
],
[
"Kara",
"Levent Burak",
""
]
] |
2504.20117 | Shubham Gandhi | Shubham Gandhi, Dhruv Shah, Manasi Patwardhan, Lovekesh Vig, Gautam
Shroff | ResearchCodeAgent: An LLM Multi-Agent System for Automated Codification
of Research Methodologies | null | AI4Research Workshop @ AAAI 2025 | null | null | cs.SE cs.AI cs.CL cs.MA | http://creativecommons.org/licenses/by/4.0/ | In this paper we introduce ResearchCodeAgent, a novel multi-agent system
leveraging large language models (LLMs) agents to automate the codification of
research methodologies described in machine learning literature. The system
bridges the gap between high-level research concepts and their practical
implementation, allowing researchers auto-generating code of existing research
papers for benchmarking or building on top-of existing methods specified in the
literature with availability of partial or complete starter code.
ResearchCodeAgent employs a flexible agent architecture with a comprehensive
action suite, enabling context-aware interactions with the research
environment. The system incorporates a dynamic planning mechanism, utilizing
both short and long-term memory to adapt its approach iteratively. We evaluate
ResearchCodeAgent on three distinct machine learning tasks with distinct task
complexity and representing different parts of the ML pipeline: data
augmentation, optimization, and data batching. Our results demonstrate the
system's effectiveness and generalizability, with 46.9% of generated code being
high-quality and error-free, and 25% showing performance improvements over
baseline implementations. Empirical analysis shows an average reduction of
57.9% in coding time compared to manual implementation. We observe higher gains
for more complex tasks. ResearchCodeAgent represents a significant step towards
automating the research implementation process, potentially accelerating the
pace of machine learning research.
| [
{
"version": "v1",
"created": "Mon, 28 Apr 2025 07:18:45 GMT"
},
{
"version": "v2",
"created": "Sat, 3 May 2025 16:45:38 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Gandhi",
"Shubham",
""
],
[
"Shah",
"Dhruv",
""
],
[
"Patwardhan",
"Manasi",
""
],
[
"Vig",
"Lovekesh",
""
],
[
"Shroff",
"Gautam",
""
]
] |
2504.20147 | Pau Amaro Seoane | Pau Amaro Seoane and Shao-Dong Zhao | A forest of gravitational waves in our Galactic Centre | Submitted. This version: Some figures updated to contain the legend
embedded in the plot | null | null | null | astro-ph.CO gr-qc | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | At the Galactic Centre, we can expect a population of a few tens of early
extreme-mass ratio inspirals (E-EMRIs) and extremely large mass ratio inspirals
(XMRIs). Depending on their evolutionary stage, they can be highly eccentric,
with moderate signal-to-noise ratios (SNRs) of tens or hundreds, or nearly
circular, with SNRs as large as a few thousand. Their individual signals
combine into a common signal, which can complicate the resolution of other
types of sources. We have calculated the foreground signal of continuous
E-EMRIs and XMRIs using a catalog based on the expected number of sources and a
realistic phase-space distribution. The forest of E-EMRIs will cover a large
portion of the LISA sensitivity curve, obscuring the signals of some massive
black hole binaries, verification binaries, and harmonics of EMRIs in their
polychromatic phase. The combined signal from XMRIs will be much weaker but
still affect intermediate-mass black hole binaries. Due to the large SNR, this
forest can be also found in other galactic nuclei, such as that of the
Andromeda galaxy. Even under conservative assumptions, the forest created by
E-EMRIs and XMRIs in our Galactic Centre will likely pose a challenge for
resolving other types of sources, as their contribution is non-coherent and
exhibits large SNRs.
| [
{
"version": "v1",
"created": "Mon, 28 Apr 2025 18:00:09 GMT"
},
{
"version": "v2",
"created": "Fri, 2 May 2025 19:30:48 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Seoane",
"Pau Amaro",
""
],
[
"Zhao",
"Shao-Dong",
""
]
] |
2504.20159 | Daniel Miller | Daniel Miller, Diego Pallero, Patricia B. Tissera, Mat\'ias Bla\~na | Differences in baryonic and dark matter scaling relations of galaxy
clusters: A comparison between IllustrisTNG simulations and observations | Accepted for publication in A&A; 17 pages, 7 figures, 5 tables, 1
appendix | null | null | null | astro-ph.CO | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We compare the self-similar baryonic mass fraction scaling relations between
galaxy clusters from the South Pole Telescope Sunyaev-Zel'dovich (SPT-SZ)
survey and the IllustrisTNG state-of-the-art magnetohydrodynamical cosmological
simulations. Using samples of 218 (TNG100) and 1605 (TNG300) friends-of-friends
(FoF) haloes within $0.0 \leq z \leq 1.5$ and $M_{200c} \geq 7 \times 10^{13}
M_{\odot}$, we fit the scaling relations using Simple Power Law (SPL), Broken
Power Law (BPL), and General Double Power Law (GDPL) models through non-linear
least squares regression. The SPL model reveals null slopes for the baryonic
fraction as a function of redshift, consistent with self-similarity.
Observations and simulations agree within $1{-}2\sigma$, suggesting comparable
baryonic scaling slopes. We identify $\sim$13.8$-$14.1 per cent of baryons as
"missing", primarily in the form of intracluster light (ICL) across all halo
masses and warm gas in low-mass haloes. High-mass haloes
($\log_{10}(M_{500c}/M_{\odot}) \geq 14$) adhere to self-similarity, while
low-mass haloes exhibit deviations, with the breakpoint occurring at
$\log_{10}(M_{500c}/M_{\odot}) \sim 14$, where baryons are redistributed to the
outskirts. Our findings suggest that the undetected warm-hot intergalactic
medium (WHIM) and baryon redistribution by feedback mechanisms are
complementary solutions to the "missing baryon" problem.
| [
{
"version": "v1",
"created": "Mon, 28 Apr 2025 18:02:42 GMT"
},
{
"version": "v2",
"created": "Fri, 2 May 2025 22:08:45 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Miller",
"Daniel",
""
],
[
"Pallero",
"Diego",
""
],
[
"Tissera",
"Patricia B.",
""
],
[
"Blaña",
"Matías",
""
]
] |
2504.20208 | Hubert Jozwiak | Hubert J\'o\'zwiak and Jaromir Tosiek | The $2$ -- D free particle in the phase space quantum mechanics | null | null | null | null | quant-ph | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The Wigner function of a free quantum particle propagating on a plane is
derived.
Two options are discussed. First the particle with fixed components of the
momentum is mentioned. Then the particle of given energy and angular momentum
is considered. The analysis does not refer to the Hilbert space formulation of
the problem at all. One starts from
a choice of coordinates on the symplectic space $(\mathbb{R}^{4},\,\omega)$
suitable for representation of eigenstates of the discussed particle. Then the
Moyal $\star_{(\text{M})}$ -- product on the phase space is derived with the
use of the Fedosov algorithm adapted to curvilinear coordinates on a flat phase
space. Next the eigenvalue equations for the Wigner eigenfunction are solved.
Finally the physically acceptable solutions are identified.
| [
{
"version": "v1",
"created": "Mon, 28 Apr 2025 19:21:27 GMT"
},
{
"version": "v2",
"created": "Fri, 2 May 2025 19:39:53 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Jóźwiak",
"Hubert",
""
],
[
"Tosiek",
"Jaromir",
""
]
] |
2504.20252 | Nibedita Padhi | Aritra Banerjee, Adrita Chakraborty, Nibedita Padhi | Non-relativistic Strings: Classical solutions and exactly solvable
models | 34+1 pages, 4 figures, comments added, references added | null | null | null | hep-th | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We discuss classical closed string solutions in non-relativistic two-sphere
target spaces. These classes of solutions closely relate to the GKP-type,
spinning and pulsating strings for the relativistic case. We derive the string
dynamics in each case and construct relevant dispersion relations, both from
the string Newton-Cartan intrinsic sigma model and using a large speed of light
expansion of relativistic Polyakov action. We further discuss construction and
characteristics of exactly solvable Neumann-Rosochatius-like dynamical systems
corresponding to strings in leading and subleading orders of the expanded
Polyakov theory.
| [
{
"version": "v1",
"created": "Mon, 28 Apr 2025 20:43:55 GMT"
},
{
"version": "v2",
"created": "Sun, 4 May 2025 18:08:11 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Banerjee",
"Aritra",
""
],
[
"Chakraborty",
"Adrita",
""
],
[
"Padhi",
"Nibedita",
""
]
] |
2504.20283 | A Singh Dr | Aman Singh | Horizon-Driven Expansion from Hawking-Like Radiation: A
Curvature-Coupled Cosmological Model | null | null | null | null | gr-qc hep-th | http://creativecommons.org/licenses/by/4.0/ | We propose a cosmological model in which the expansion of the universe is
driven by a Hawking-like influx of energy across the cosmological horizon,
rather than from a fixed cosmological constant. In place of a cosmological
constant, we introduce source terms in the Friedmann and continuity equations
that couple horizon curvature to matter and radiation densities. At high
curvature (large Hubble parameter $H$), this influx strongly replenishes matter
and radiation, slowing their adiabatic dilution. As curvature diminishes, the
influx weakens, smoothly transitioning into standard radiation- or
matter-dominated eras. This mechanism naturally suppresses spatial curvature
without requiring an inflationary phase. It may also produce
near-scale-invariant fluctuations via slowly varying horizon thermodynamics.
| [
{
"version": "v1",
"created": "Mon, 28 Apr 2025 21:53:17 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Singh",
"Aman",
""
]
] |
2504.20304 | Xiulin Yang | Xiulin Yang, Zhuoxuan Ju, Lanni Bu, Zoey Liu, Nathan Schneider | UD-English-CHILDES: A Collected Resource of Gold and Silver Universal
Dependencies Trees for Child Language Interactions | null | null | null | null | cs.CL cs.AI | http://creativecommons.org/licenses/by/4.0/ | CHILDES is a widely used resource of transcribed child and child-directed
speech. This paper introduces UD-English-CHILDES, the first officially released
Universal Dependencies (UD) treebank derived from previously
dependency-annotated CHILDES data with consistent and unified annotation
guidelines. Our corpus harmonizes annotations from 11 children and their
caregivers, totaling over 48k sentences. We validate existing gold-standard
annotations under the UD v2 framework and provide an additional 1M
silver-standard sentences, offering a consistent resource for computational and
linguistic research.
| [
{
"version": "v1",
"created": "Mon, 28 Apr 2025 23:20:36 GMT"
},
{
"version": "v2",
"created": "Sun, 4 May 2025 13:35:09 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Yang",
"Xiulin",
""
],
[
"Ju",
"Zhuoxuan",
""
],
[
"Bu",
"Lanni",
""
],
[
"Liu",
"Zoey",
""
],
[
"Schneider",
"Nathan",
""
]
] |
2504.20335 | Bowen Jiang | Bowen Jiang and Chaofan Ma and Duo Wang | VA-CDH: A Variance-Aware Method to Optimize Latency for Caching with
Delayed Hits | null | null | null | null | cs.NI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Caches are fundamental to latency-sensitive systems like Content Delivery
Networks (CDNs) and Mobile Edge Computing (MEC). However, the delayed hit
phenomenon where multiple requests for an object occur during its fetch from
the remote server after a miss significantly inflates user-perceived latency.
While recent algorithms acknowledge delayed hits by estimating the resulting
aggregate delay, they predominantly focus on its mean value. We identify and
demonstrate that such approaches are insufficient, as the real aggregate delay
frequently exhibits substantial variance in the true production system, leading
to suboptimal latency performance when ignored. Thus, we propose VA-CDH, a
variance-aware method to optimize latency for caching with delayed hits. It
employs a novel ranking function that explicitly incorporates both the
empirically estimated mean and standard deviation of aggregate delay, allowing
caching decisions to account for its variation. We derive the analytical
distribution of aggregate delay under Poisson arrivals as a theoretical
contribution, offering more statistical insight beyond the mean value. Through
the simulations conducted on synthetic and real-world datasets, we show that
VA-CDH reduces the total latency by 1%-6% approximately compared to
state-of-the-art algorithms.
| [
{
"version": "v1",
"created": "Tue, 29 Apr 2025 00:58:59 GMT"
},
{
"version": "v2",
"created": "Sat, 3 May 2025 01:10:30 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Jiang",
"Bowen",
""
],
[
"Ma",
"Chaofan",
""
],
[
"Wang",
"Duo",
""
]
] |
2504.20341 | LiFang Chen | Li-Fang Chen | Preliminary Design and Performance Simulation of a Thermal Neutron
Diffractometer Using McStas | 20 pages, 4 figures, preliminary technical report on McStas-based
simulation input for thermal neutron diffractometers, submitted for academic
sharing via arXiv only | null | null | null | physics.ins-det | http://creativecommons.org/licenses/by/4.0/ | This study presents a preliminary design and simulation of a thermal neutron
diffractometer using the McStas Monte Carlo ray-tracing package. The simulated
system is based on generalized instrument parameters, including typical
collimator divergences, monochromator mosaic spreads, and detector positioning,
rather than on any specific engineering blueprint. The simulation evaluates key
performance metrics such as neutron flux distribution, beam divergence, and
wavelength resolution along the beam path.
Results demonstrate that, even under simplified geometric assumptions, the
model can provide insight into component alignment, resolution limitations, and
flux attenuation behavior. The modular structure of the simulation allows easy
parameter adjustments for different instrument configurations and serves as a
foundation for more advanced system-specific optimization studies.
Additionally, the provided McStas input file serves as a reusable and
customizable simulation template that can significantly reduce modeling time
for engineering personnel during the early-stage design and layout phase. This
tool also facilitates technical training and comparative analysis without
relying on finalized engineering drawings.
| [
{
"version": "v1",
"created": "Tue, 29 Apr 2025 01:21:25 GMT"
},
{
"version": "v2",
"created": "Mon, 5 May 2025 02:50:48 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Chen",
"Li-Fang",
""
]
] |
2504.20344 | Chaohan Cui | Chaohan Cui, Prajit Dhara, and Saikat Guha | Coherent State Assisted Entanglement Generation Between Quantum Memories | 17 pages, 5 figures | null | null | null | quant-ph | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Generating entanglement deterministically at a capacity-approaching rate is
critical for next-generation quantum networks. We propose
weak-coherent-state-assisted protocols that can generate entanglement
near-deterministically between reflective-cavity-based quantum memories at a
success rate that exceeds the 50\% limit associated with single-photon-mediated
schemes. The most pronounced benefit is shown in the low-channel-loss regime
and persists even with moderate noise. We extend our protocols to entangle an
array of memories in a GHZ state, and infer that it yields an exponential
speed-up compared to previous single-photon-based protocols.
| [
{
"version": "v1",
"created": "Tue, 29 Apr 2025 01:27:32 GMT"
},
{
"version": "v2",
"created": "Sat, 3 May 2025 18:34:08 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Cui",
"Chaohan",
""
],
[
"Dhara",
"Prajit",
""
],
[
"Guha",
"Saikat",
""
]
] |
2504.20347 | Grigoris Panotopoulos | Grigoris Panotopoulos | Radial Oscillations of the HESS J1731-347 Compact Object via the
Karmarkar Condition in Gravity | References added | Universe 2025, 11(5), 146 | 10.3390/universe11050146 | null | gr-qc astro-ph.HE astro-ph.SR | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We model the light HESS J1731-347 compact object (of known stellar mass and
radius) within Einstein's General Relativity imposing the Karmarkar condition
in gravity for anisotropic stars. The three free parameters of the analytic
solution are determined imposing the matching conditions at the surface of the
star for objects of known stellar mass and radius. Finally, using well
established criteria it is shown that the solution is compatible with all
requirements for well behaved and realistic solutions. Furthermore, we study
the radial oscillation modes, and we compare to the ones corresponding to an
isotropic star modeled by the Tolman IV exact analytic solution obtained long
time ago. A comparison between the large frequency separations is made as well.
| [
{
"version": "v1",
"created": "Tue, 29 Apr 2025 01:31:29 GMT"
},
{
"version": "v2",
"created": "Sat, 3 May 2025 21:52:01 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Panotopoulos",
"Grigoris",
""
]
] |
2504.20348 | Varatheepan Paramanayakam | Varatheepan Paramanayakam, Andreas Karatzas, Iraklis Anagnostopoulos,
Dimitrios Stamoulis | CarbonCall: Sustainability-Aware Function Calling for Large Language
Models on Edge Devices | null | null | null | null | cs.PF cs.AI cs.SY eess.SY | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Large Language Models (LLMs) enable real-time function calling in edge AI
systems but introduce significant computational overhead, leading to high power
consumption and carbon emissions. Existing methods optimize for performance
while neglecting sustainability, making them inefficient for energy-constrained
environments. We introduce CarbonCall, a sustainability-aware function-calling
framework that integrates dynamic tool selection, carbon-aware execution, and
quantized LLM adaptation. CarbonCall adjusts power thresholds based on
real-time carbon intensity forecasts and switches between model variants to
sustain high tokens-per-second throughput under power constraints. Experiments
on an NVIDIA Jetson AGX Orin show that CarbonCall reduces carbon emissions by
up to 52%, power consumption by 30%, and execution time by 30%, while
maintaining high efficiency.
| [
{
"version": "v1",
"created": "Tue, 29 Apr 2025 01:37:08 GMT"
},
{
"version": "v2",
"created": "Fri, 2 May 2025 19:16:14 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Paramanayakam",
"Varatheepan",
""
],
[
"Karatzas",
"Andreas",
""
],
[
"Anagnostopoulos",
"Iraklis",
""
],
[
"Stamoulis",
"Dimitrios",
""
]
] |
2504.20372 | Prasanna Ramakrishnan | Moses Charikar, Prasanna Ramakrishnan, Kangning Wang | Approximately Dominating Sets in Elections | null | null | null | null | cs.GT cs.DM cs.DS math.CO | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Condorcet's paradox is a fundamental result in social choice theory which
states that there exist elections in which, no matter which candidate wins, a
majority of voters prefer a different candidate. In fact, even if we can select
any $k$ winners, there still may exist another candidate that would beat each
of the winners in a majority vote. That is, elections may require arbitrarily
large dominating sets.
We show that approximately dominating sets of constant size always exist. In
particular, for every $\varepsilon > 0$, every election (irrespective of the
number of voters or candidates) can select $O(\frac{1}{\varepsilon ^2})$
winners such that no other candidate beats each of the winners by a margin of
more than $\varepsilon$ fraction of voters.
Our proof uses a simple probabilistic construction using samples from a
maximal lottery, a well-studied distribution over candidates derived from the
Nash equilibrium of a two-player game. In stark contrast to general approximate
equilibria, which may require support logarithmic in the number of pure
strategies, we show that maximal lotteries can be approximated with constant
support size. These approximate maximal lotteries may be of independent
interest.
| [
{
"version": "v1",
"created": "Tue, 29 Apr 2025 02:30:00 GMT"
},
{
"version": "v2",
"created": "Mon, 5 May 2025 03:55:02 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Charikar",
"Moses",
""
],
[
"Ramakrishnan",
"Prasanna",
""
],
[
"Wang",
"Kangning",
""
]
] |
2504.20466 | Yi Yang Woo | Woo Yi Yang, Jiarui Wang, Sijing Wu, Huiyu Duan, Yuxin Zhu, Liu Yang,
Kang Fu, Guangtao Zhai, Xiongkuo Min | LMME3DHF: Benchmarking and Evaluating Multimodal 3D Human Face
Generation with LMMs | null | null | null | null | cs.CV | http://creativecommons.org/licenses/by/4.0/ | The rapid advancement in generative artificial intelligence have enabled the
creation of 3D human faces (HFs) for applications including media production,
virtual reality, security, healthcare, and game development, etc. However,
assessing the quality and realism of these AI-generated 3D human faces remains
a significant challenge due to the subjective nature of human perception and
innate perceptual sensitivity to facial features. To this end, we conduct a
comprehensive study on the quality assessment of AI-generated 3D human faces.
We first introduce Gen3DHF, a large-scale benchmark comprising 2,000 videos of
AI-Generated 3D Human Faces along with 4,000 Mean Opinion Scores (MOS)
collected across two dimensions, i.e., quality and authenticity, 2,000
distortion-aware saliency maps and distortion descriptions. Based on Gen3DHF,
we propose LMME3DHF, a Large Multimodal Model (LMM)-based metric for Evaluating
3DHF capable of quality and authenticity score prediction, distortion-aware
visual question answering, and distortion-aware saliency prediction.
Experimental results show that LMME3DHF achieves state-of-the-art performance,
surpassing existing methods in both accurately predicting quality scores for
AI-generated 3D human faces and effectively identifying distortion-aware
salient regions and distortion types, while maintaining strong alignment with
human perceptual judgments. Both the Gen3DHF database and the LMME3DHF will be
released upon the publication.
| [
{
"version": "v1",
"created": "Tue, 29 Apr 2025 07:00:06 GMT"
},
{
"version": "v2",
"created": "Mon, 5 May 2025 16:07:30 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Yang",
"Woo Yi",
""
],
[
"Wang",
"Jiarui",
""
],
[
"Wu",
"Sijing",
""
],
[
"Duan",
"Huiyu",
""
],
[
"Zhu",
"Yuxin",
""
],
[
"Yang",
"Liu",
""
],
[
"Fu",
"Kang",
""
],
[
"Zhai",
"Guangtao",
""
],
[
"Min",
"Xiongkuo",
""
]
] |
2504.20512 | Ahmed Bayoumi | Ahmed Bayoumi, Ahmed Khalil, Pol Van Dorpe, Maumita Chakrabarti,
Dimitrios Velenis, Philippe Absil, Filippo Ferraro, Yoojin Ban, Joris Van
Campenhout, Wim Bogaerts, Qingzhong Deng | Broadband and Ultra-Compact Adiabatic Coupler Based on Linearly Tapered
Silicon Waveguides | null | null | null | null | physics.optics physics.app-ph | http://creativecommons.org/licenses/by-nc-nd/4.0/ | 2x2 couplers are crucial in many photonics applications but are often
wavelength-sensitive. Alternative designs could be large, complex, or lossy. We
present a broadband ultra-compact adiabatic coupler with minimal coupling
variation and the least silicon taper length of 1.44 um, to our best knowledge.
| [
{
"version": "v1",
"created": "Tue, 29 Apr 2025 07:52:16 GMT"
},
{
"version": "v2",
"created": "Wed, 30 Apr 2025 09:31:39 GMT"
},
{
"version": "v3",
"created": "Sat, 3 May 2025 18:27:57 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Bayoumi",
"Ahmed",
""
],
[
"Khalil",
"Ahmed",
""
],
[
"Van Dorpe",
"Pol",
""
],
[
"Chakrabarti",
"Maumita",
""
],
[
"Velenis",
"Dimitrios",
""
],
[
"Absil",
"Philippe",
""
],
[
"Ferraro",
"Filippo",
""
],
[
"Ban",
"Yoojin",
""
],
[
"Van Campenhout",
"Joris",
""
],
[
"Bogaerts",
"Wim",
""
],
[
"Deng",
"Qingzhong",
""
]
] |
2504.20543 | Khun Sang Phukon | Khun Sang Phukon, Nathan K. Johnson-McDaniel, Amitesh Singh, Anuradha
Gupta | Evolution of precessing binary black holes on eccentric orbits using
orbit-averaged evolution equations | Update with a reduced-size version of Figure 3, and authors' email
addresses and ORCID details | null | null | LIGO-P2500139 | gr-qc astro-ph.HE | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The most general bound binary black hole (BBH) system has an eccentric orbit
and precessing spins. The detection of such a system with significant
eccentricity close to the merger would be a clear signature of dynamical
formation. In order to study such systems, it is important to be able to evolve
their spins and eccentricity from the larger separations at which the binary
formed to the smaller separations at which it is detected, or vice versa.
Knowledge of the precessional evolution of the binary's orbital angular
momentum can also be used to twist up aligned-spin eccentric waveform models to
create a spin-precessing eccentric waveform model. In this paper, we present a
new publicly available code to evolve eccentric, precessing BBHs using
orbit-averaged post-Newtonian (PN) equations from the literature. The
spin-precession dynamics is 2PN accurate, i.e., with the leading spin-orbit and
spin-spin corrections. The evolution of orbital parameters (orbital frequency,
eccentricity, and periastron precession), which follow the quasi-Keplerian
parametrization, is 3PN accurate in the point particle terms and includes the
leading order spin-orbit and spin-spin effects. All the spin-spin terms include
the quadrupole-monopole interaction. The eccentricity enhancement functions in
the fluxes use the high-accuracy hyperasymptotic expansions from Loutrel and
Yunes [Classical Quantum Gravity {\bf 34} 044003 (2017)]. We discuss various
features of the code and study the evolution of the orbital and spin-precession
parameters of eccentric, precessing BBHs. In particular, we study the
dependence of the spin morphologies on eccentricity, where we find that the
transition point from one spin morphology to another can depend
nonmonotonically on eccentricity, and the fraction of binaries in a given
morphology at a given point in the evolution of a population depends on the
instantaneous eccentricity.
| [
{
"version": "v1",
"created": "Tue, 29 Apr 2025 08:40:07 GMT"
},
{
"version": "v2",
"created": "Mon, 5 May 2025 11:55:56 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Phukon",
"Khun Sang",
""
],
[
"Johnson-McDaniel",
"Nathan K.",
""
],
[
"Singh",
"Amitesh",
""
],
[
"Gupta",
"Anuradha",
""
]
] |
2504.20682 | Long Liu | Long Liu, Cihui Yang | OG-HFYOLO :Orientation gradient guidance and heterogeneous feature
fusion for deformation table cell instance segmentation | null | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Table structure recognition is a key task in document analysis. However, the
geometric deformation in deformed tables causes a weak correlation between
content information and structure, resulting in downstream tasks not being able
to obtain accurate content information. To obtain fine-grained spatial
coordinates of cells, we propose the OG-HFYOLO model, which enhances the edge
response by Gradient Orientation-aware Extractor, combines a Heterogeneous
Kernel Cross Fusion module and a scale-aware loss function to adapt to
multi-scale objective features, and introduces mask-driven non-maximal
suppression in the post-processing, which replaces the traditional bounding box
suppression mechanism. Furthermore, we also propose a data generator, filling
the gap in the dataset for fine-grained deformation table cell spatial
coordinate localization, and derive a large-scale dataset named Deformation
Wired Table (DWTAL). Experiments show that our proposed model demonstrates
excellent segmentation accuracy on all mainstream instance segmentation models.
The dataset and the source code are open source:
https://github.com/justliulong/OGHFYOLO.
| [
{
"version": "v1",
"created": "Tue, 29 Apr 2025 12:02:01 GMT"
},
{
"version": "v2",
"created": "Sat, 3 May 2025 13:15:33 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Liu",
"Long",
""
],
[
"Yang",
"Cihui",
""
]
] |
2504.20834 | Alan Lee | Alan Lee and Harry Tong | Token-Efficient RL for LLM Reasoning | Title updated to "Token-Efficient RL for LLM Reasoning" to better
reflect algorithmic focus. Revised abstract, intro, and conclusion. Paper
shortened and typos fixed | null | null | null | cs.LG cs.AI | http://creativecommons.org/licenses/by/4.0/ | We propose reinforcement learning (RL) strategies tailored for reasoning in
large language models (LLMs) under strict memory and compute limits, with a
particular focus on compatibility with LoRA fine-tuning. Rather than relying on
full-sequence updates or separate critic networks, we design critic-free
methods that operate on a small, informative subset of output tokens to reduce
memory usage and stabilize training. We introduce S-GRPO, a stochastic variant
of Group Relative Policy Optimization, and T-SPMO, a token-level prefix
matching approach for fine-grained credit assignment. Applied to Qwen2-1.5B,
our methods raise accuracy on the SVAMP benchmark from 46% to over 70% and show
strong performance on multi-digit multiplication. Surprisingly, full-token GRPO
under LoRA fails to improve over the base model, suggesting that selective
token-level optimization may act as an implicit regularizer in low-parameter
training regimes.
| [
{
"version": "v1",
"created": "Tue, 29 Apr 2025 14:58:43 GMT"
},
{
"version": "v2",
"created": "Mon, 5 May 2025 15:36:15 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Lee",
"Alan",
""
],
[
"Tong",
"Harry",
""
]
] |
2504.20898 | Hasan Md Tusfiqur Alam | Hasan Md Tusfiqur Alam, Devansh Srivastav, Abdulrahman Mohamed Selim,
Md Abdul Kadir, Md Moktadirul Hoque Shuvo, Daniel Sonntag | CBM-RAG: Demonstrating Enhanced Interpretability in Radiology Report
Generation with Multi-Agent RAG and Concept Bottleneck Models | Accepted in the 17th ACM SIGCHI Symposium on Engineering Interactive
Computing Systems (EICS 2025) | null | 10.1145/3731406.3731970 | null | cs.AI cs.CV cs.IR | http://creativecommons.org/licenses/by-nc-sa/4.0/ | Advancements in generative Artificial Intelligence (AI) hold great promise
for automating radiology workflows, yet challenges in interpretability and
reliability hinder clinical adoption. This paper presents an automated
radiology report generation framework that combines Concept Bottleneck Models
(CBMs) with a Multi-Agent Retrieval-Augmented Generation (RAG) system to bridge
AI performance with clinical explainability. CBMs map chest X-ray features to
human-understandable clinical concepts, enabling transparent disease
classification. Meanwhile, the RAG system integrates multi-agent collaboration
and external knowledge to produce contextually rich, evidence-based reports.
Our demonstration showcases the system's ability to deliver interpretable
predictions, mitigate hallucinations, and generate high-quality, tailored
reports with an interactive interface addressing accuracy, trust, and usability
challenges. This framework provides a pathway to improving diagnostic
consistency and empowering radiologists with actionable insights.
| [
{
"version": "v1",
"created": "Tue, 29 Apr 2025 16:14:55 GMT"
},
{
"version": "v2",
"created": "Sun, 4 May 2025 22:38:21 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Alam",
"Hasan Md Tusfiqur",
""
],
[
"Srivastav",
"Devansh",
""
],
[
"Selim",
"Abdulrahman Mohamed",
""
],
[
"Kadir",
"Md Abdul",
""
],
[
"Shuvo",
"Md Moktadirul Hoque",
""
],
[
"Sonntag",
"Daniel",
""
]
] |
2504.20903 | Sai Mihir Jakkaraju | Prothit Sen, Sai Mihir Jakkaraju | Modeling AI-Human Collaboration as a Multi-Agent Adaptation | null | null | null | null | cs.MA cs.AI cs.HC | http://creativecommons.org/licenses/by/4.0/ | We develop an agent-based simulation to formalize AI-human collaboration as a
function of task structure, advancing a generalizable framework for strategic
decision-making in organizations. Distinguishing between heuristic-based human
adaptation and rule-based AI search, we model interactions across modular
(parallel) and sequenced (interdependent) tasks using an NK model. Our results
reveal that in modular tasks, AI often substitutes for humans - delivering
higher payoffs unless human expertise is very high, and the AI search space is
either narrowly focused or extremely broad. In sequenced tasks, interesting
complementarities emerge. When an expert human initiates the search and AI
subsequently refines it, aggregate performance is maximized. Conversely, when
AI leads, excessive heuristic refinement by the human can reduce payoffs. We
also show that even "hallucinatory" AI - lacking memory or structure - can
improve outcomes when augmenting low-capability humans by helping escape local
optima. These results yield a robust implication: the effectiveness of AI-human
collaboration depends less on context or industry, and more on the underlying
task structure. By elevating task decomposition as the central unit of
analysis, our model provides a transferable lens for strategic decision-making
involving humans and an agentic AI across diverse organizational settings.
| [
{
"version": "v1",
"created": "Tue, 29 Apr 2025 16:19:53 GMT"
},
{
"version": "v2",
"created": "Mon, 5 May 2025 04:26:02 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Sen",
"Prothit",
""
],
[
"Jakkaraju",
"Sai Mihir",
""
]
] |
2504.20955 | Elias Mann | Elias L. Mann, Corin C. Wagen, Jonathon E. Vandezande, Arien M. Wagen,
Spencer C. Schneider | Egret-1: Pretrained Neural Network Potentials For Efficient and Accurate
Bioorganic Simulation | 33 pages, 8 figures | null | null | null | physics.chem-ph | http://creativecommons.org/licenses/by/4.0/ | Accurate simulation of atomic systems has the potential to revolutionize the
design of molecules and materials. Unfortunately, exact solutions of the
Schr\"odinger equation scale as O(N!) and remain inaccessible for systems with
more than a handful of atoms, forcing scientists to accept steep tradeoffs
between speed and accuracy and limiting the reliability and utility of the
resultant simulations. Recent work in machine learning has demonstrated that
neural network potentials (NNPs) can learn efficient approximations to quantum
mechanics and resolve this tradeoff, but existing NNPs still suffer from
limited accuracy relative to state-of-the-art quantum-chemical methods. Here,
we present Egret-1, a family of large pretrained NNPs based on the MACE
architecture with general applicability to main-group, organic, and
biomolecular chemistry. We find that the Egret-1 models equal or exceed the
accuracy of routinely employed quantum-chemical methods on a variety of
standard tasks, including torsional scans, conformer ranking, and geometry
optimization, while offering multiple-order-of-magnitude speedups relative to
legacy methods. We also highlight important lacunae for future NNP research to
investigate, and suggest strategies for building future high-quality models
with increased scale and generality.
| [
{
"version": "v1",
"created": "Tue, 29 Apr 2025 17:26:21 GMT"
},
{
"version": "v2",
"created": "Wed, 30 Apr 2025 15:29:40 GMT"
},
{
"version": "v3",
"created": "Thu, 1 May 2025 21:15:45 GMT"
},
{
"version": "v4",
"created": "Mon, 5 May 2025 15:18:40 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Mann",
"Elias L.",
""
],
[
"Wagen",
"Corin C.",
""
],
[
"Vandezande",
"Jonathon E.",
""
],
[
"Wagen",
"Arien M.",
""
],
[
"Schneider",
"Spencer C.",
""
]
] |
2504.21104 | Dmitriy Pak | Dmitriy G. Pak, Pengming Zhang, Takuya Tsukioka | Weyl multiplet structure of QCD | 9 pages, 1 figure | null | null | null | hep-th | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Weyl group symmetric structure of SU(3) quantum chromodynamics (QCD) with one
flavor quark is considered. It has been demonstrated that Weyl group as a
finite color subgroup of SU(3) provides an intrinsic color symmetry of quark
and gluon fields on mass shell. We show that standard QCD equations of motion
lead to systems of equations for quarks and gluons forming non-trivial
irreducible multiplets in one- and two-dimensional representations of the Weyl
group. Some implications of Weyl multiplet structure in QCD and hadron physics
are considered.
| [
{
"version": "v1",
"created": "Tue, 29 Apr 2025 18:26:03 GMT"
},
{
"version": "v2",
"created": "Mon, 5 May 2025 16:49:08 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Pak",
"Dmitriy G.",
""
],
[
"Zhang",
"Pengming",
""
],
[
"Tsukioka",
"Takuya",
""
]
] |
2504.21214 | Jinzhao Zhou | Jinzhao Zhou, Zehong Cao, Yiqun Duan, Connor Barkley, Daniel Leong,
Xiaowei Jiang, Quoc-Toan Nguyen, Ziyi Zhao, Thomas Do, Yu-Cheng Chang,
Sheng-Fu Liang, Chin-teng Lin | Pretraining Large Brain Language Model for Active BCI: Silent Speech | null | null | null | null | cs.CL cs.AI eess.AS | http://creativecommons.org/licenses/by/4.0/ | This paper explores silent speech decoding in active brain-computer interface
(BCI) systems, which offer more natural and flexible communication than
traditional BCI applications. We collected a new silent speech dataset of over
120 hours of electroencephalogram (EEG) recordings from 12 subjects, capturing
24 commonly used English words for language model pretraining and decoding.
Following the recent success of pretraining large models with self-supervised
paradigms to enhance EEG classification performance, we propose Large Brain
Language Model (LBLM) pretrained to decode silent speech for active BCI. To
pretrain LBLM, we propose Future Spectro-Temporal Prediction (FSTP) pretraining
paradigm to learn effective representations from unlabeled EEG data. Unlike
existing EEG pretraining methods that mainly follow a masked-reconstruction
paradigm, our proposed FSTP method employs autoregressive modeling in temporal
and frequency domains to capture both temporal and spectral dependencies from
EEG signals. After pretraining, we finetune our LBLM on downstream tasks,
including word-level and semantic-level classification. Extensive experiments
demonstrate significant performance gains of the LBLM over fully-supervised and
pretrained baseline models. For instance, in the difficult cross-session
setting, our model achieves 47.0\% accuracy on semantic-level classification
and 39.6\% in word-level classification, outperforming baseline methods by
5.4\% and 7.3\%, respectively. Our research advances silent speech decoding in
active BCI systems, offering an innovative solution for EEG language model
pretraining and a new dataset for fundamental research.
| [
{
"version": "v1",
"created": "Tue, 29 Apr 2025 22:48:27 GMT"
},
{
"version": "v2",
"created": "Sat, 3 May 2025 07:53:27 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Zhou",
"Jinzhao",
""
],
[
"Cao",
"Zehong",
""
],
[
"Duan",
"Yiqun",
""
],
[
"Barkley",
"Connor",
""
],
[
"Leong",
"Daniel",
""
],
[
"Jiang",
"Xiaowei",
""
],
[
"Nguyen",
"Quoc-Toan",
""
],
[
"Zhao",
"Ziyi",
""
],
[
"Do",
"Thomas",
""
],
[
"Chang",
"Yu-Cheng",
""
],
[
"Liang",
"Sheng-Fu",
""
],
[
"Lin",
"Chin-teng",
""
]
] |
2504.21496 | Samuel Corson | Samuel M. Corson | Bounded displacement permutations on tree-like spaces | null | null | null | null | math.GR math.CO | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | It is shown that if a metric space exhibits certain finiteness and tree-like
properties, then elements of its group of bounded displacement which are
infinitely divisible are also torsion. This extends a result of N. M. Suchkov,
A. A. Shlepkin, and D. A. Taysnyov.
| [
{
"version": "v1",
"created": "Wed, 30 Apr 2025 10:30:33 GMT"
},
{
"version": "v2",
"created": "Mon, 5 May 2025 07:09:43 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Corson",
"Samuel M.",
""
]
] |
2504.21560 | Gonzalo Herrera | Gonzalo Herrera | Plausible Indication of Gamma-Ray Absorption by Dark Matter in NGC 1068 | 10 pages, 6 figures, added clarifications and citations | null | null | null | hep-ph astro-ph.CO astro-ph.HE | http://creativecommons.org/licenses/by/4.0/ | NGC 1068 is the brightest extragalactic source in high-energy neutrinos as
seen by IceCube, yet the accompanying gamma-ray flux is orders of magnitude
weaker. It has been argued that this indicates that the bulk of neutrinos and
gamma rays are emitted in the innermost vicinity of the central supermassive
black hole, which is transparent to neutrinos, but opaque to gamma rays. Even
in such extreme scenarios for the acceleration of cosmic rays, astrophysical
models typically overestimate the low-energy gamma-ray flux and/or require some
fine-tuning in the physical parameters. Here we suggest instead that the dark
matter surrounding the supermassive black hole may absorb the gamma rays,
inducing the observed deficit. We show that for a dark matter-photon scattering
cross section in the range $\sigma_{\rm DM-\gamma}/m_{\rm DM} \simeq
10^{-28}-10^{-30}$ cm$^2$/GeV, Fermi-LAT measurements can be well reconciled
with IceCube data. We also present some simple particle physics examples that
achieve the correct spectral energy dependence while respecting complementary
constraints.
| [
{
"version": "v1",
"created": "Wed, 30 Apr 2025 11:59:02 GMT"
},
{
"version": "v2",
"created": "Sun, 4 May 2025 15:58:47 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Herrera",
"Gonzalo",
""
]
] |
2504.21581 | Yuxin Jing | Yuxin Jing, Yuchen Zheng, Jufeng Zhao, Guangmang Cui, Tianpei Zhang | Make Both Ends Meet: A Synergistic Optimization Infrared Small Target
Detection with Streamlined Computational Overhead | null | null | null | null | eess.IV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Infrared small target detection(IRSTD) is widely recognized as a challenging
task due to the inherent limitations of infrared imaging, including low
signal-to-noise ratios, lack of texture details, and complex background
interference. While most existing methods model IRSTD as a semantic
segmentation task, but they suffer from two critical drawbacks: (1)blurred
target boundaries caused by long-distance imaging dispersion; and (2) excessive
computational overhead due to indiscriminate feature stackin. To address these
issues, we propose the Lightweight Efficiency Infrared Small Target Detection
(LE-IRSTD), a lightweight and efficient framework based on YOLOv8n, with
following key innovations. Firstly, we identify that the multiple bottleneck
structures within the C2f component of the YOLOv8-n backbone contribute to an
increased computational burden. Therefore, we implement the Mobile Inverted
Bottleneck Convolution block (MBConvblock) and Bottleneck Structure block
(BSblock) in the backbone, effectively balancing the trade-off between
computational efficiency and the extraction of deep semantic information.
Secondly, we introduce the Attention-based Variable Convolution Stem (AVCStem)
structure, substituting the final convolution with Variable Kernel Convolution
(VKConv), which allows for adaptive convolutional kernels that can transform
into various shapes, facilitating the receptive field for the extraction of
targets. Finally, we employ Global Shuffle Convolution (GSConv) to shuffle the
channel dimension features obtained from different convolutional approaches,
thereby enhancing the robustness and generalization capabilities of our method.
Experimental results demonstrate that our LE-IRSTD method achieves compelling
results in both accuracy and lightweight performance, outperforming several
state-of-the-art deep learning methods.
| [
{
"version": "v1",
"created": "Wed, 30 Apr 2025 12:38:42 GMT"
},
{
"version": "v2",
"created": "Sun, 4 May 2025 13:20:32 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Jing",
"Yuxin",
""
],
[
"Zheng",
"Yuchen",
""
],
[
"Zhao",
"Jufeng",
""
],
[
"Cui",
"Guangmang",
""
],
[
"Zhang",
"Tianpei",
""
]
] |
2504.21601 | Fernando A N Santos | Danillo Barros de Souza, Jonatas Teodomiro, Fernando A. N. Santos,
Mengjun Ding, Weiqiang Sun, Mathieu Desroches, J\"urgen Jost and Serafim
Rodrigues | Efficient Decomposition of Forman-Ricci Curvature on Vietoris-Rips
Complexes and Data Applications | null | null | null | null | math.GT cs.CG cs.DM cs.DS math.CO | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Discrete Forman-Ricci curvature (FRC) is an efficient tool that characterizes
essential geometrical features and associated transitions of real-world
networks, extending seamlessly to higher-dimensional computations in simplicial
complexes. In this article, we provide two major advancements: First, we give a
decomposition for FRC, enabling local computations of FRC. Second, we construct
a set-theoretical proof enabling an efficient algorithm for the local
computation of FRC in Vietoris-Rips (VR) complexes.Strikingly, this approach
reveals critical information and geometric insights often overlooked by
conventional classification techniques. Our findings open new avenues for
geometric computations in VR complexes and highlight an essential yet
under-explored aspect of data classification: the geometry underpinning
statistical patterns.
| [
{
"version": "v1",
"created": "Wed, 30 Apr 2025 12:59:23 GMT"
},
{
"version": "v2",
"created": "Mon, 5 May 2025 08:53:57 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"de Souza",
"Danillo Barros",
""
],
[
"Teodomiro",
"Jonatas",
""
],
[
"Santos",
"Fernando A. N.",
""
],
[
"Ding",
"Mengjun",
""
],
[
"Sun",
"Weiqiang",
""
],
[
"Desroches",
"Mathieu",
""
],
[
"Jost",
"Jürgen",
""
],
[
"Rodrigues",
"Serafim",
""
]
] |
2504.21638 | Owen Ekblad | Owen Ekblad | A note on the quantum Wielandt inequality | 7 pages; the author was supported in part by US National Science
Foundation Grant No. DMS-2153946; comments welcome; previous version had a
minor error in Proposition 2.2 that has now been fixed, and the presentation
has been updated accordingly | null | null | null | quant-ph cs.IT math.IT math.OA | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | In this note, we prove that the index of primitivity of any primitive unital
Schwarz map is at most $2(D-1)^2$, where $D$ is the dimension of the underlying
matrix algebra. This inequality was first proved by Rahaman for Schwarz maps
which were both unital and trace preserving. As we show, the assumption of
unitality is basically innocuous, but in general not all primitive unital
Schwarz maps are trace preserving. Therefore, the precise purpose of this note
is to showcase how to apply the method of Rahaman to unital primitive Schwarz
maps that don't preserve trace. As a corollary of this theorem, we show that
the index of primitivity of any primitive 2-positive map is at most $2(D-1)^2$,
so in particular this bound holds for arbitrary primitive completely positive
maps. We briefly discuss of how this relates to a conjecture of Perez-Garcia,
Verstraete, Wolf and Cirac.
| [
{
"version": "v1",
"created": "Wed, 30 Apr 2025 13:40:53 GMT"
},
{
"version": "v2",
"created": "Fri, 2 May 2025 20:36:20 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Ekblad",
"Owen",
""
]
] |
2504.21693 | Zhu Longkang | Longkang Zhu, Xinli Shi, Xiangping Xu, Jinde Cao | Distributed Online Randomized Gradient-Free optimization with Compressed
Communication | We are withdrawing this manuscript due to factual errors identified
in all the remarks presented. We will revise and resubmit a corrected version
in the future | null | null | null | math.OC | http://creativecommons.org/licenses/by/4.0/ | This paper addresses two fundamental challenges in distributed online convex
optimization: communication efficiency and optimization under limited feedback.
We propose Online Compressed Gradient Tracking with one-point Bandit Feedback
(OCGT-BF), a novel algorithm that harness data compression and gradient-free
optimization techniques in distributed networks. Our algorithm incorporates a
compression scheme with error compensation mechanisms to reduce communication
overhead while maintaining convergence guarantees. Unlike traditional
approaches that assume perfect communication and full gradient access, OCGT-BF
operates effectively under practical constraints by combining gradient-like
tracking with one-point feedback estimation. We provide theoretical analysis
demonstrating the dynamic regret bounds under both bandit feedback and
stochastic gradient scenarios. Finally, extensive experiments validate that
OCGT-BF achieves low dynamic regret while significantly reducing communication
requirements.
| [
{
"version": "v1",
"created": "Wed, 30 Apr 2025 14:33:47 GMT"
},
{
"version": "v2",
"created": "Sun, 4 May 2025 15:43:11 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Zhu",
"Longkang",
""
],
[
"Shi",
"Xinli",
""
],
[
"Xu",
"Xiangping",
""
],
[
"Cao",
"Jinde",
""
]
] |
2504.21760 | Seyed Amin Seyed Fakhari | Takayuki Hibi, Seyed Amin Seyed Fakhari | Bounded powers of edge ideals: Gorenstein toric rings | null | null | null | null | math.AC math.CO | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Let $S=K[x_1, \ldots,x_n]$ denote the polynomial ring in $n$ variables over a
field $K$ and $I \subset S$ a monomial ideal. Given a vector
$\mathfrak{c}\in\mathbb{N}^n$, the ideal $I_{\mathfrak{c}}$ is the ideal
generated by those monomials belonging to $I$ whose exponent vectors are
componentwise bounded above by $\mathfrak{c}$. Let $\delta_{\mathfrak{c}}(I)$
be the largest integer $q$ for which $(I^q)_{\mathfrak{c}}\neq 0$. For a finite
graph $G$, its edge ideal is denoted by $I(G)$. Let
$\mathcal{B}(\mathfrak{c},G)$ be the toric ring which is generated by the
monomials belonging to the minimal system of monomial generators of
$(I(G)^{\delta_{\mathfrak{c}}(I)})_{\mathfrak{c}}$. In a previous work, the
authors proved that $(I(G)^{\delta_{\mathfrak{c}}(I)})_{\mathfrak{c}}$ is a
polymatroidal ideal. It follows that $\mathcal{B}(\mathfrak{c},G)$ is a normal
Cohen--Macaulay domain. In this paper, we study the Gorenstein property of
$\mathcal{B}(\mathfrak{c},G)$.
| [
{
"version": "v1",
"created": "Wed, 30 Apr 2025 16:00:19 GMT"
},
{
"version": "v2",
"created": "Sun, 4 May 2025 19:46:00 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Hibi",
"Takayuki",
""
],
[
"Fakhari",
"Seyed Amin Seyed",
""
]
] |
2504.21772 | Eunil Park | Minwoo Oh, Minsu Park, Eunil Park | Solving Copyright Infringement on Short Video Platforms: Novel Datasets
and an Audio Restoration Deep Learning Pipeline | will be presented in IJCAI 2025, 9 pages, 4 tables, 3 figures | null | null | null | cs.MM cs.AI | http://creativecommons.org/licenses/by-nc-sa/4.0/ | Short video platforms like YouTube Shorts and TikTok face significant
copyright compliance challenges, as infringers frequently embed arbitrary
background music (BGM) to obscure original soundtracks (OST) and evade content
originality detection. To tackle this issue, we propose a novel pipeline that
integrates Music Source Separation (MSS) and cross-modal video-music retrieval
(CMVMR). Our approach effectively separates arbitrary BGM from the original
OST, enabling the restoration of authentic video audio tracks. To support this
work, we introduce two domain-specific datasets: OASD-20K for audio separation
and OSVAR-160 for pipeline evaluation. OASD-20K contains 20,000 audio clips
featuring mixed BGM and OST pairs, while OSVAR160 is a unique benchmark dataset
comprising 1,121 video and mixed-audio pairs, specifically designed for short
video restoration tasks. Experimental results demonstrate that our pipeline not
only removes arbitrary BGM with high accuracy but also restores OSTs, ensuring
content integrity. This approach provides an ethical and scalable solution to
copyright challenges in user-generated content on short video platforms.
| [
{
"version": "v1",
"created": "Wed, 30 Apr 2025 16:17:05 GMT"
},
{
"version": "v2",
"created": "Sat, 3 May 2025 12:54:39 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Oh",
"Minwoo",
""
],
[
"Park",
"Minsu",
""
],
[
"Park",
"Eunil",
""
]
] |
2504.21795 | Yuankang Zhao | Yuankang Zhao, Matthew Engelhard | Balancing Interpretability and Flexibility in Modeling Diagnostic
Trajectories with an Embedded Neural Hawkes Process Model | null | null | null | null | stat.ML cs.LG stat.AP | http://creativecommons.org/licenses/by-nc-sa/4.0/ | The Hawkes process (HP) is commonly used to model event sequences with
self-reinforcing dynamics, including electronic health records (EHRs).
Traditional HPs capture self-reinforcement via parametric impact functions that
can be inspected to understand how each event modulates the intensity of
others. Neural network-based HPs offer greater flexibility, resulting in
improved fit and prediction performance, but at the cost of interpretability,
which is often critical in healthcare. In this work, we aim to understand and
improve upon this tradeoff. We propose a novel HP formulation in which impact
functions are modeled by defining a flexible impact kernel, instantiated as a
neural network, in event embedding space, which allows us to model large-scale
event sequences with many event types. This approach is more flexible than
traditional HPs yet more interpretable than other neural network approaches,
and allows us to explicitly trade flexibility for interpretability by adding
transformer encoder layers to further contextualize the event embeddings.
Results show that our method accurately recovers impact functions in
simulations, achieves competitive performance on MIMIC-IV procedure dataset,
and gains clinically meaningful interpretation on XX-EHR with children
diagnosis dataset even without transformer layers. This suggests that our
flexible impact kernel is often sufficient to capture self-reinforcing dynamics
in EHRs and other data effectively, implying that interpretability can be
maintained without loss of performance.
| [
{
"version": "v1",
"created": "Wed, 30 Apr 2025 16:52:43 GMT"
},
{
"version": "v2",
"created": "Sat, 3 May 2025 19:17:50 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Zhao",
"Yuankang",
""
],
[
"Engelhard",
"Matthew",
""
]
] |
2505.00001 | Michael Lu | Shaun Baek, Shaun Esua-Mensah, Cyrus Tsui, Sejan Vigneswaralingam,
Abdullah Alali, Michael Lu, Vasu Sharma, Sean O'Brien, Kevin Zhu | Rosetta-PL: Propositional Logic as a Benchmark for Large Language Model
Reasoning | null | null | null | null | cs.CL | http://creativecommons.org/licenses/by/4.0/ | Large Language Models (LLMs) are primarily trained on high-resource natural
languages, limiting their effectiveness in low-resource settings and in tasks
requiring deep logical reasoning. This research introduces Rosetta-PL, a
benchmark designed to evaluate LLMs' logical reasoning and generalization
capabilities in a controlled environment. We construct Rosetta-PL by
translating a dataset of logical propositions from Lean into a custom logical
language, which is then used to fine-tune an LLM (e.g., GPT-4o). Our
experiments analyze the impact of the size of the dataset and the translation
methodology on the performance of the model. Our results indicate that
preserving logical relationships in the translation process significantly
boosts precision, with accuracy plateauing beyond roughly 20,000 training
samples. These insights provide valuable guidelines for optimizing LLM training
in formal reasoning tasks and improving performance in various low-resource
language applications.
| [
{
"version": "v1",
"created": "Tue, 25 Mar 2025 21:12:29 GMT"
},
{
"version": "v2",
"created": "Fri, 2 May 2025 17:47:00 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Baek",
"Shaun",
""
],
[
"Esua-Mensah",
"Shaun",
""
],
[
"Tsui",
"Cyrus",
""
],
[
"Vigneswaralingam",
"Sejan",
""
],
[
"Alali",
"Abdullah",
""
],
[
"Lu",
"Michael",
""
],
[
"Sharma",
"Vasu",
""
],
[
"O'Brien",
"Sean",
""
],
[
"Zhu",
"Kevin",
""
]
] |
2505.00034 | Zijie Lin | Zijie Lin, Zikang Liu, Hanbo Fan | Improving Phishing Email Detection Performance of Small Large Language
Models | null | null | null | null | cs.CL cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Large language models(LLMs) have demonstrated remarkable performance on many
natural language processing(NLP) tasks and have been employed in phishing email
detection research. However, in current studies, well-performing LLMs typically
contain billions or even tens of billions of parameters, requiring enormous
computational resources. To reduce computational costs, we investigated the
effectiveness of small-parameter LLMs for phishing email detection. These LLMs
have around 3 billion parameters and can run on consumer-grade GPUs. However,
small LLMs often perform poorly in phishing email detection task. To address
these issues, we designed a set of methods including Prompt Engineering,
Explanation Augmented Fine-tuning, and Model Ensemble to improve phishing email
detection capabilities of small LLMs. We validated the effectiveness of our
approach through experiments, significantly improving both accuracy and F1
score on the SpamAssassin and CEAS\_08 datasets. Furthermore, the fine-tuned
models demonstrated strong transferability, achieving robust performance across
multiple unseen phishing datasets, outperforming traditional baselines and
approaching standard-sized LLMs.
| [
{
"version": "v1",
"created": "Tue, 29 Apr 2025 14:07:06 GMT"
},
{
"version": "v2",
"created": "Sat, 3 May 2025 13:08:45 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Lin",
"Zijie",
""
],
[
"Liu",
"Zikang",
""
],
[
"Fan",
"Hanbo",
""
]
] |
2505.00041 | Ritik Raj | Ritik Raj, Shengjie Lin, William Won, Tushar Krishna | MCMComm: Hardware-Software Co-Optimization for End-to-End Communication
in Multi-Chip-Modules | null | null | null | null | cs.AR | http://creativecommons.org/licenses/by/4.0/ | Increasing AI computing demands and slowing transistor scaling have led to
the advent of Multi-Chip-Module (MCMs) based accelerators. MCMs enable
cost-effective scalability, higher yield, and modular reuse by partitioning
large chips into smaller chiplets. However, MCMs come at an increased
communication cost, which requires critical analysis and optimization. This
paper makes three main contributions: (i) an end-to-end, off-chip
congestion-aware and packaging-adaptive analytical framework for detailed
analysis, (ii) hardware software co-optimization incorporating diagonal links,
on-chip redistribution, and non-uniform workload partitioning to optimize the
framework, and (iii) using metaheuristics (genetic algorithms, GA) and mixed
integer quadratic programming (MIQP) to solve the optimized framework.
Experimental results demonstrate significant performance improvements for CNNs
and Vision Transformers, showcasing up to 1.58x and 2.7x EdP (Energy delay
Product) improvement using GA and MIQP, respectively.
| [
{
"version": "v1",
"created": "Tue, 29 Apr 2025 20:06:57 GMT"
},
{
"version": "v2",
"created": "Fri, 2 May 2025 19:27:43 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Raj",
"Ritik",
""
],
[
"Lin",
"Shengjie",
""
],
[
"Won",
"William",
""
],
[
"Krishna",
"Tushar",
""
]
] |
2505.00091 | Tengchao Zhang | Tengchao Zhang, Yonglin Tian, Fei Lin, Jun Huang, Patrik P. S\"uli,
Rui Qin and Fei-Yue Wang | CoordField: Coordination Field for Agentic UAV Task Allocation In
Low-altitude Urban Scenarios | Submitted ITSC 2025 | null | null | null | cs.RO cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | With the increasing demand for heterogeneous Unmanned Aerial Vehicle (UAV)
swarms to perform complex tasks in urban environments, system design now faces
major challenges, including efficient semantic understanding, flexible task
planning, and the ability to dynamically adjust coordination strategies in
response to evolving environmental conditions and continuously changing task
requirements. To address the limitations of existing approaches, this paper
proposes coordination field agentic system for coordinating heterogeneous UAV
swarms in complex urban scenarios. In this system, large language models (LLMs)
is responsible for interpreting high-level human instructions and converting
them into executable commands for the UAV swarms, such as patrol and target
tracking. Subsequently, a Coordination field mechanism is proposed to guide UAV
motion and task selection, enabling decentralized and adaptive allocation of
emergent tasks. A total of 50 rounds of comparative testing were conducted
across different models in a 2D simulation space to evaluate their performance.
Experimental results demonstrate that the proposed system achieves superior
performance in terms of task coverage, response time, and adaptability to
dynamic changes.
| [
{
"version": "v1",
"created": "Wed, 30 Apr 2025 18:02:45 GMT"
},
{
"version": "v2",
"created": "Sat, 3 May 2025 16:45:20 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Zhang",
"Tengchao",
""
],
[
"Tian",
"Yonglin",
""
],
[
"Lin",
"Fei",
""
],
[
"Huang",
"Jun",
""
],
[
"Süli",
"Patrik P.",
""
],
[
"Qin",
"Rui",
""
],
[
"Wang",
"Fei-Yue",
""
]
] |
2505.00203 | Gabriele Grillo | Dorothea-Enrica von Criegern, Gabriele Grillo, Dario Monticelli | Nonexistence of Solutions to classes of parabolic inequalities in the
Riemannian setting | null | null | null | null | math.AP math.DG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We establish conditions for nonexistence of global solutions for a class of
quasilinear parabolic problems with a potential on complete, non-compact
Riemannian manifolds, including the Porous Medium Equation and the p-Laplacian
with a potential term. Our results reveal the interplay between the manifold's
geometry, the power nonlinearity, and the potential's behavior at infinity.
Using a test function argument, we identify explicit parameter ranges where
nonexistence holds.
| [
{
"version": "v1",
"created": "Wed, 30 Apr 2025 22:03:15 GMT"
},
{
"version": "v2",
"created": "Mon, 5 May 2025 09:39:02 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"von Criegern",
"Dorothea-Enrica",
""
],
[
"Grillo",
"Gabriele",
""
],
[
"Monticelli",
"Dario",
""
]
] |
2505.00241 | Takara Taniguchi | Takara Taniguchi, Ryo Ohashi, Tsuyoshi Takagi | Enumeration Algorithm for Genus-4 Superspecial Hyperelliptic Curves with
Automorphism Group $Q_8$ | null | null | null | null | math.AG | http://creativecommons.org/licenses/by/4.0/ | In this paper, we propose an algorithm to enumerate genus-4 superspecial
hyperelliptic curves whose automorphism groups isomorphic to the quaternion
group. By implementing this algorithm with Magma, we successfully obtain the
number of isomorphism classes of such curves in every characteristic $7 \leq p
< 10000$. Interestingly, the experimental results lead us to the conjecture
that there exist exactly $[p/48]$ isomorphism classes of such curves if $p
\equiv 1,7 \pmod{8}$, whereas such curves exist if $p \equiv 3,5 \pmod{8}$
| [
{
"version": "v1",
"created": "Thu, 1 May 2025 01:27:01 GMT"
},
{
"version": "v2",
"created": "Sat, 3 May 2025 23:24:48 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Taniguchi",
"Takara",
""
],
[
"Ohashi",
"Ryo",
""
],
[
"Takagi",
"Tsuyoshi",
""
]
] |
2505.00300 | Luciano Combi | Luciano Combi, Christopher Thompson, Daniel M. Siegel, Alexander
Philippov, Bart Ripperda | Magnetized Accretion onto Neutron Stars: from Photon-trapped to
Neutrino-cooled Flows | 27 pages, 23 figures, 6 appendices. v1: fixed typo in caption and
acknowledgement | null | null | null | astro-ph.HE gr-qc | http://creativecommons.org/licenses/by/4.0/ | When a neutron star (NS) intercepts gas from a non-degenerate star, e.g., in
a tidal disruption event, a common-envelope phase, or the collapsing core of a
massive star, photons become trapped in the hot flow around the NS. This gas
forms a radiatively inefficient accretion flow (RIAF) until the density and
temperature close to the NS surface grow large enough for binding energy to be
converted to neutrinos. Here we present three-dimensional,
general-relativistic, magnetohydrodynamic simulations of accretion onto a
non-rotating, unmagnetized NS. These connect, for the first time, an extended
accretion disk with a self-consistent hydrostatic atmosphere around the star.
The impact of different seed magnetic fields and accretion rates is studied by
approximating the radiation-pressure dominated flow as an ideal gas with an
adiabatic index of $4/3$, coupled to a variable neutrino emissivity. At low
accretion rates, the hydrostatic atmosphere shows slow rotation and weak
magnetization, transitioning to an outer RIAF structure. A toroidal magnetic
field mediates the inward flow of energy and angular momentum through the
atmosphere, which reaches a steady state when neutrino emission balances the
accretion power. We develop a one-dimensional analytical model connecting these
results with more general initial conditions and describing the main features
of the flow. Our results have implications for the spin and mass evolution of
hypercritically accreting NSs.
| [
{
"version": "v1",
"created": "Thu, 1 May 2025 04:44:24 GMT"
},
{
"version": "v2",
"created": "Sat, 3 May 2025 02:54:09 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Combi",
"Luciano",
""
],
[
"Thompson",
"Christopher",
""
],
[
"Siegel",
"Daniel M.",
""
],
[
"Philippov",
"Alexander",
""
],
[
"Ripperda",
"Bart",
""
]
] |
2505.00316 | Tien Comlekoglu | Tien Comlekoglu, J. Quetzalc\'oatl Toledo-Mar\'in, Tina Comlekoglu,
Douglas W. DeSimone, Shayn M. Peirce, Geoffrey Fox, James A. Glazier | Surrogate modeling of Cellular-Potts Agent-Based Models as a
segmentation task using the U-Net neural network architecture | null | null | null | null | cs.LG cs.AI q-bio.QM | http://creativecommons.org/licenses/by/4.0/ | The Cellular-Potts model is a powerful and ubiquitous framework for
developing computational models for simulating complex multicellular biological
systems. Cellular-Potts models (CPMs) are often computationally expensive due
to the explicit modeling of interactions among large numbers of individual
model agents and diffusive fields described by partial differential equations
(PDEs). In this work, we develop a convolutional neural network (CNN) surrogate
model using a U-Net architecture that accounts for periodic boundary
conditions. We use this model to accelerate the evaluation of a mechanistic CPM
previously used to investigate \textit{in vitro} vasculogenesis. The surrogate
model was trained to predict 100 computational steps ahead (Monte-Carlo steps,
MCS), accelerating simulation evaluations by a factor of 590 times compared to
CPM code execution. Over multiple recursive evaluations, our model effectively
captures the emergent behaviors demonstrated by the original Cellular-Potts
model of such as vessel sprouting, extension and anastomosis, and contraction
of vascular lacunae. This approach demonstrates the potential for deep learning
to serve as efficient surrogate models for CPM simulations, enabling faster
evaluation of computationally expensive CPM of biological processes at greater
spatial and temporal scales.
| [
{
"version": "v1",
"created": "Thu, 1 May 2025 05:30:38 GMT"
},
{
"version": "v2",
"created": "Mon, 5 May 2025 15:26:29 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Comlekoglu",
"Tien",
""
],
[
"Toledo-Marín",
"J. Quetzalcóatl",
""
],
[
"Comlekoglu",
"Tina",
""
],
[
"DeSimone",
"Douglas W.",
""
],
[
"Peirce",
"Shayn M.",
""
],
[
"Fox",
"Geoffrey",
""
],
[
"Glazier",
"James A.",
""
]
] |
2505.00334 | Luigi Sigillo | Luigi Sigillo, Christian Bianchi, Aurelio Uncini, Danilo Comminiello | Quaternion Wavelet-Conditioned Diffusion Models for Image
Super-Resolution | Accepted for presentation at IJCNN 2025 | null | null | null | cs.CV cs.LG | http://creativecommons.org/licenses/by/4.0/ | Image Super-Resolution is a fundamental problem in computer vision with broad
applications spacing from medical imaging to satellite analysis. The ability to
reconstruct high-resolution images from low-resolution inputs is crucial for
enhancing downstream tasks such as object detection and segmentation. While
deep learning has significantly advanced SR, achieving high-quality
reconstructions with fine-grained details and realistic textures remains
challenging, particularly at high upscaling factors. Recent approaches
leveraging diffusion models have demonstrated promising results, yet they often
struggle to balance perceptual quality with structural fidelity. In this work,
we introduce ResQu a novel SR framework that integrates a quaternion wavelet
preprocessing framework with latent diffusion models, incorporating a new
quaternion wavelet- and time-aware encoder. Unlike prior methods that simply
apply wavelet transforms within diffusion models, our approach enhances the
conditioning process by exploiting quaternion wavelet embeddings, which are
dynamically integrated at different stages of denoising. Furthermore, we also
leverage the generative priors of foundation models such as Stable Diffusion.
Extensive experiments on domain-specific datasets demonstrate that our method
achieves outstanding SR results, outperforming in many cases existing
approaches in perceptual quality and standard evaluation metrics. The code will
be available after the revision process.
| [
{
"version": "v1",
"created": "Thu, 1 May 2025 06:17:33 GMT"
},
{
"version": "v2",
"created": "Mon, 5 May 2025 10:20:23 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Sigillo",
"Luigi",
""
],
[
"Bianchi",
"Christian",
""
],
[
"Uncini",
"Aurelio",
""
],
[
"Comminiello",
"Danilo",
""
]
] |
2505.00372 | Nicola Branchini | Nicola Branchini, V\'ictor Elvira | Towards Adaptive Self-Normalized Importance Samplers | Accepted at the 2025 IEEE Statistical Signal Processing Workshop;
fixed a few maths typos and comments about RIS | null | null | null | stat.CO | http://creativecommons.org/licenses/by/4.0/ | The self-normalized importance sampling (SNIS) estimator is a Monte Carlo
estimator widely used to approximate expectations in statistical signal
processing and machine learning.
The efficiency of SNIS depends on the choice of proposal, but selecting a
good proposal is typically unfeasible. In particular, most of the existing
adaptive IS (AIS) literature overlooks the optimal SNIS proposal.
In this paper, we introduce an AIS framework that uses MCMC to approximate
the optimal SNIS proposal within an iterative scheme. This is, to the best of
our knowledge, the first AIS framework targeting specifically the SNIS optimal
proposal. We find a close connection with adaptive schemes used in ratio
importance sampling (RIS), which also brings a new perspective and paves the
way for combining techniques from AIS and adaptive RIS. We outline possible
extensions, connections with existing MCMC-driven AIS algorithms, theoretical
directions, and demonstrate performance in numerical examples.
| [
{
"version": "v1",
"created": "Thu, 1 May 2025 07:56:02 GMT"
},
{
"version": "v2",
"created": "Sun, 4 May 2025 12:27:26 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Branchini",
"Nicola",
""
],
[
"Elvira",
"Víctor",
""
]
] |
2505.00411 | Yang Liu | Yang Liu | Affine matrix scrambling achieves smoothness-dependent convergence rates | The author has identified significant errors in the proofs of Lemma 2
and Lemma 4. Consequently, the main results of this manuscript cannot be
substantiated. The author plans to revise the work to pursue weaker
theoretical guarantees and address these issues | null | null | null | math.NA cs.NA | http://creativecommons.org/licenses/by/4.0/ | We study the convergence rate of the median estimator for affine matrix
scrambled digital nets applied to integrands over the unit hypercube $[0,
1]^s$. By taking the median of $(2r-1)$ independent randomized quasi-Monte
Carlo (RQMC) samples, we demonstrate that the desired convergence rates can be
achieved without increasing the number of randomizations $r$ as the quadrature
size $N$ grows for both bounded and unbounded integrands. For unbounded
integrands, our analysis assumes a boundary growth condition on the weak
derivatives and also considers singularities such as kinks and jump
discontinuities. Notably, when $r = 1$, the median estimator reduces to the
standard RQMC estimator. By applying analytical techniques developed for median
estimators, we prove that the affine matrix scrambled estimator achieves a
convergence rate depending on the integrand's smoothness, and is therefore not
limited by the canonical rate $\mathcal{O}(N^{-3/2})$. However, this
smoothness-dependent theoretical rate is not observed empirically in numerical
experiments when the affine matrix scrambling yields a heavy-tailed sampling
distribution. In contrast, the median estimator consistently reveals the
theoretical rates and yields smaller integration errors than mean estimators,
further highlighting its advantages.
| [
{
"version": "v1",
"created": "Thu, 1 May 2025 09:09:58 GMT"
},
{
"version": "v2",
"created": "Sat, 3 May 2025 17:37:19 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Liu",
"Yang",
""
]
] |
2505.00420 | Elio Marconi | Alberto Bressan, Elio Marconi, Ganesh Vaidya | Uniqueness Domains for ${\bf L}^\infty$ Solutions of $2 \times 2$
Hyperbolic Conservation Laws | 24 pages, 12 figures | null | null | null | math.AP | http://creativecommons.org/licenses/by/4.0/ | For a genuinely nonlinear $2\times 2$ hyperbolic system of conservation laws,
assuming that the initial data have small ${\bf L}^\infty$ norm but possibly
unbounded total variation, the existence of global solutions was proved in a
classical paper by Glimm and Lax (1970). In general, the total variation of
these solutions decays like $t^{-1}$. Motivated by the theory of fractional
domains for linear analytic semigroups, we consider here solutions with faster
decay rate: $\mathrm{Tot. Var. }\bigl\{u(t,\cdot)\bigr\}\leq C t^{\alpha-1}$.
For these solutions, a uniqueness theorem is proved. Indeed, as the initial
data range over a domain of functions with $\|\bar u\|_{{\bf L}^\infty}
\leq\varepsilon_1$ small enough, solutions with fast decay yield a H\"older
continuous semigroup. The H\"older exponent can be taken arbitrarily close to
$1$ by further shrinking the value of $\varepsilon_1>0$. An auxiliary result
identifies a class of initial data whose solutions have rapidly decaying total
variation.
| [
{
"version": "v1",
"created": "Thu, 1 May 2025 09:35:54 GMT"
},
{
"version": "v2",
"created": "Sat, 3 May 2025 12:19:09 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Bressan",
"Alberto",
""
],
[
"Marconi",
"Elio",
""
],
[
"Vaidya",
"Ganesh",
""
]
] |
2505.00449 | Bart Jacobs | Bart Jacobs and Justus Fasse | An approach for modularly verifying the core of Rust's atomic reference
counting algorithm against the (X)C20 memory consistency model | 21 pages, 5 figures | null | null | null | cs.PL | http://creativecommons.org/licenses/by/4.0/ | We propose an approach for modular verification of programs that use
relaxed-consistency atomic memory access primitives and fences, sufficient for
verifying the core of Rust's Atomic Reference Counting (ARC) algorithm, and we
argue its soundness, when combined with a simple static analysis and admitting
an open sub-problem, with respect to the C20 memory consistency model, as well
as, even in the absence of any static analysis and without any assumptions,
with respect to XC20, a recently proposed minor strengthening of C20 that rules
out out-of-thin-air behaviors but allows load buffering. In contrast to
existing work on verifying ARC, we do not assume acyclicity of the union of the
program-order and reads-from relations. We define an interleaving operational
semantics, prove its soundness with respect to (X)C20's axiomatic semantics,
and then apply any existing program logic for fine-grained interleaving
concurrency, such as Iris.
| [
{
"version": "v1",
"created": "Thu, 1 May 2025 10:51:20 GMT"
},
{
"version": "v2",
"created": "Fri, 2 May 2025 10:35:22 GMT"
},
{
"version": "v3",
"created": "Mon, 5 May 2025 09:37:23 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Jacobs",
"Bart",
""
],
[
"Fasse",
"Justus",
""
]
] |
2505.00468 | Youngkyu Kim | Youngkyu Kim, Byounghyun Yoo, Ji Young Yun, Hyeokmin Lee, Sehyeon
Park, Jin Woo Moon, Eun Ji Choi | Evaluation of Thermal Control Based on Spatial Thermal Comfort with
Reconstructed Environmental Data | null | null | null | null | cs.CE | http://creativecommons.org/licenses/by/4.0/ | Achieving thermal comfort while maintaining energy efficiency is a critical
objective in building system control. Conventional thermal comfort models, such
as the Predicted Mean Vote (PMV), rely on both environmental and personal
variables. However, the use of fixed-location sensors limits the ability to
capture spatial variability, which reduces the accuracy of occupant-specific
comfort estimation. To address this limitation, this study proposes a new PMV
estimation method that incorporates spatial environmental data reconstructed
using the Gappy Proper Orthogonal Decomposition (Gappy POD) algorithm. In
addition, a group PMV-based control framework is developed to account for the
thermal comfort of multiple occupants. The Gappy POD method enables fast and
accurate reconstruction of indoor temperature fields from sparse sensor
measurements. Using these reconstructed fields and occupant location data,
spatially resolved PMV values are calculated. Group-level thermal conditions
are then derived through statistical aggregation methods and used to control
indoor temperature in a multi-occupant living lab environment. Experimental
results show that the Gappy POD algorithm achieves an average relative error
below 3\% in temperature reconstruction. PMV distributions varied by up to 1.26
scale units depending on occupant location. Moreover, thermal satisfaction
outcomes varied depending on the group PMV method employed. These findings
underscore the importance for adaptive thermal control strategies that
incorporate both spatial and individual variability, offering valuable insights
for future occupant-centric building operations.
| [
{
"version": "v1",
"created": "Thu, 1 May 2025 11:43:28 GMT"
},
{
"version": "v2",
"created": "Sun, 4 May 2025 07:08:44 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Kim",
"Youngkyu",
""
],
[
"Yoo",
"Byounghyun",
""
],
[
"Yun",
"Ji Young",
""
],
[
"Lee",
"Hyeokmin",
""
],
[
"Park",
"Sehyeon",
""
],
[
"Moon",
"Jin Woo",
""
],
[
"Choi",
"Eun Ji",
""
]
] |
2505.00475 | Ni Liu | Ni Liu and J.-Q. Liang (Institute of Theoretical Physics, State Key
Laboratory of Quantum Optics and Quantum Optics Devices, Shanxi University) | Quantum mechanics of inverted potential well -- Hermitian Hamiltonian
with imaginary eigenvalues, quantum-classical correspondence | null | null | null | null | quant-ph | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We in this paper study the quantization of a particle in an inverted
potential well. The Hamiltonian is Hermitian, while the potential is unbounded
below. Classically the particle moves away acceleratingly from the center of
potential top. The existing eigenstates must be unstable with imaginary
eigenvalues, which characterize the decay rate of states. We solve the
Hamiltonian problem of inverted potential well by the algebraic method with
imaginary-frequency raising and lowering boson operators similar to the normal
oscillator case. The boson number operator is non-Hermitian, while the
integer-number eigenvalues are, of course, real. Dual sets of eigenstates,
denoted by "bra" and "ket", are requested corresponding respectively to the
complex conjugate number-operators. Orthonormal condition exists between the
"bra" and "ket" states. We derive a spatially non-localized generating
function, from which $n$-th eigenfunctions can be generated by the raising
operators in coordinate representation. The "bra" and "ket" generating
functions are mutually normalized with the imaginary integration measure. The
probability density operators defined between the "bra" and "ket" states are
non-Hermitian invariants, which lead to the Schr\H{o}dinger equations
respectively for the "bra" and "ket" states. While probabilities of "bra" and
"ket" states themselves are not conserved quantities because of the decay. The
imaginary-frequency boson coherent states are defined as eigenstates of
lowering operators. The minimum uncertainty relation is proved explicitly in
the coherent states. Finally the probability average of Heisenberg equation in
the coherent states is shown precisely in agreement with the classical equation
of motion. The quantum-classical correspondence exists in the imaginary
eigenvalue system.
| [
{
"version": "v1",
"created": "Thu, 1 May 2025 11:59:33 GMT"
},
{
"version": "v2",
"created": "Mon, 5 May 2025 10:03:23 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Liu",
"Ni",
"",
"Institute of Theoretical Physics, State Key\n Laboratory of Quantum Optics and Quantum Optics Devices, Shanxi University"
],
[
"Liang",
"J. -Q.",
"",
"Institute of Theoretical Physics, State Key\n Laboratory of Quantum Optics and Quantum Optics Devices, Shanxi University"
]
] |
2505.00503 | Ke Jiang | Ke Jiang and Wen Jiang and Masahiro Fujisawa and Xiaoyang Tan | Variational OOD State Correction for Offline Reinforcement Learning | null | null | null | null | cs.LG cs.AI cs.RO | http://creativecommons.org/licenses/by-nc-nd/4.0/ | The performance of Offline reinforcement learning is significantly impacted
by the issue of state distributional shift, and out-of-distribution (OOD) state
correction is a popular approach to address this problem. In this paper, we
propose a novel method named Density-Aware Safety Perception (DASP) for OOD
state correction. Specifically, our method encourages the agent to prioritize
actions that lead to outcomes with higher data density, thereby promoting its
operation within or the return to in-distribution (safe) regions. To achieve
this, we optimize the objective within a variational framework that
concurrently considers both the potential outcomes of decision-making and their
density, thus providing crucial contextual information for safe
decision-making. Finally, we validate the effectiveness and feasibility of our
proposed method through extensive experimental evaluations on the offline
MuJoCo and AntMaze suites.
| [
{
"version": "v1",
"created": "Thu, 1 May 2025 13:14:07 GMT"
},
{
"version": "v2",
"created": "Mon, 5 May 2025 06:00:10 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Jiang",
"Ke",
""
],
[
"Jiang",
"Wen",
""
],
[
"Fujisawa",
"Masahiro",
""
],
[
"Tan",
"Xiaoyang",
""
]
] |
2505.00516 | Nicholas Latham | K. Abe, S. Abe, R. Akutsu, H. Alarakia-Charles, Y. I. Alj Hakim, S.
Alonso Monsalve, L. Anthony, S. Aoki, K. A. Apte, T. Arai, T. Arihara, S.
Arimoto, E. T. Atkin, N. Babu, V. Baranov, G. J. Barker, G. Barr, D. Barrow,
P. Bates, L. Bathe-Peters, M. Batkiewicz-Kwasniak, N. Baudis, V. Berardi, L.
Berns, S. Bhattacharjee, A. Blanchet, A. Blondel, P. M. M. Boistier, S.
Bolognesi, S. Bordoni, S. B. Boyd, C. Bronner, A. Bubak, M. Buizza Avanzini,
J. A. Caballero, F. Cadoux, N. F. Calabria, S. Cao, S. Cap, D. Carabadjac, S.
L. Cartwright, M. P. Casado, M. G. Catanesi, J. Chakrani, A. Chalumeau, D.
Cherdack, A. Chvirova, J. Coleman, G. Collazuol, F. Cormier, A. A. L.
Craplet, A. Cudd, D. D'ago, C. Dalmazzone, T. Daret, P. Dasgupta, C. Davis,
Yu. I. Davydov, P. de Perio, G. De Rosa, T. Dealtry, C. Densham, A.
Dergacheva, R. Dharmapal Banerjee, F. Di Lodovico, G. Diaz Lopez, S. Dolan,
D. Douqa, T. A. Doyle, O. Drapier, K. E. Duffy, J. Dumarchez, P. Dunne, K.
Dygnarowicz, A. Eguchi, J. Elias, S. Emery-Schrenk, G. Erofeev, A. Ershova,
G. Eurin, D. Fedorova, S. Fedotov, M. Feltre, L. Feng, D. Ferlewicz, A. J.
Finch, M. D. Fitton, C. Forza, M. Friend, Y. Fujii, Y. Fukuda, Y. Furui, J.
Garc\'ia-Marcos, A. C. Germer, L. Giannessi, C. Giganti, M. Girgus, V.
Glagolev, M. Gonin, R. Gonz\'alez Jim\'enez, J. Gonz\'alez Rosa, E. A. G.
Goodman, K. Gorshanov, P. Govindaraj, M. Grassi, M. Guigue, F. Y. Guo, D. R.
Hadley, S. Han, D. A. Harris, R. J. Harris, T. Hasegawa, C. M. Hasnip, S.
Hassani, N. C. Hastings, Y. Hayato, I. Heitkamp, D. Henaff, Y. Hino, J.
Holeczek, A. Holin, T. Holvey, N. T. Hong Van, T. Honjo, M. C. F. Hooft, K.
Hosokawa, J. Hu, A. K. Ichikawa, K. Ieki, M. Ikeda, T. Ishida, M. Ishitsuka,
H. Ito, S. Ito, A. Izmaylov, N. Jachowicz, S. J. Jenkins, C. Jes\'us-Valls,
M. Jia, J. J. Jiang, J. Y. Ji, T. P. Jones, P. Jonsson, S. Joshi, M.
Kabirnezhad, A. C. Kaboth, H. Kakuno, J. Kameda, S. Karpova, V. S. Kasturi,
Y. Kataoka, T. Katori, A. Kawabata, Y. Kawamura, M. Kawaue, E. Kearns, M.
Khabibullin, A. Khotjantsev, T. Kikawa, S. King, V. Kiseeva, J. Kisiel, A.
Klustov\'a, L. Kneale, H. Kobayashi, L. Koch, S. Kodama, M. Kolupanova, A.
Konaka, L. L. Kormos, Y. Koshio, K. Kowalik, Y. Kudenko, Y. Kudo, A. Kumar
Jha, R. Kurjata, V. Kurochka, T. Kutter, L. Labarga, M. Lachat, K. Lachner,
J. Lagoda, S. M. Lakshmi, M. Lamers James, A. Langella, D. H. Langridge,
J.-F. Laporte, D. Last, N. Latham, M. Laveder, L. Lavitola, M. Lawe, D. Leon
Silverio, S. Levorato, S. V. Lewis, B. Li, C. Lin, R. P. Litchfield, S. L.
Liu, W. Li, A. Longhin, A. Lopez Moreno, L. Ludovici, X. Lu, T. Lux, L. N.
Machado, L. Magaletti, K. Mahn, K. K. Mahtani, S. Manly, A. D. Marino, D. G.
R. Martin, D. A. Martinez Caicedo, L. Martinez, M. Martini, T. Matsubara, R.
Matsumoto, V. Matveev, C. Mauger, K. Mavrokoridis, N. McCauley, K. S.
McFarland, C. McGrew, J. McKean, A. Mefodiev, G. D. Megias, L. Mellet, C.
Metelko, M. Mezzetto, S. Miki, V. Mikola, E. W. Miller, A. Minamino, O.
Mineev, S. Mine, J. Mirabito, M. Miura, S. Moriyama, S. Moriyama, P.
Morrison, Th. A. Mueller, D. Munford, A. Mu\~noz, L. Munteanu, Y. Nagai, T.
Nakadaira, K. Nakagiri, M. Nakahata, Y. Nakajima, K. D. Nakamura, A. Nakano,
Y. Nakano, S. Nakayama, T. Nakaya, K. Nakayoshi, C. E. R. Naseby, D. T.
Nguyen, V. Q. Nguyen, K. Niewczas, S. Nishimori, Y. Nishimura, Y. Noguchi, T.
Nosek, F. Nova, J. C. Nugent, H. M. O'Keeffe, L. O'Sullivan, R. Okazaki, W.
Okinaga, K. Okumura, T. Okusawa, N. Onda, N. Ospina, L. Osu, Y. Oyama, V.
Paolone, J. Pasternak, D. Payne, M. Pfaff, L. Pickering, B. Popov, A. J.
Portocarrero Yrey, M. Posiadala-Zezula, Y. S. Prabhu, H. Prasad, F. Pupilli,
B. Quilain, P. T. Quyen, E. Radicioni, B. Radics, M. A. Ramirez, R. Ramsden,
P. N. Ratoff, M. Reh, G. Reina, C. Riccio, D. W. Riley, E. Rondio, S. Roth,
N. Roy, A. Rubbia, L. Russo, A. Rychter, W. Saenz, K. Sakashita, S. Samani,
F. S\'anchez, E. M. Sandford, Y. Sato, T. Schefke, K. Scholberg, M. Scott, Y.
Seiya, T. Sekiguchi, H. Sekiya, T. Sekiya, D. Seppala, D. Sgalaberna, A.
Shaikhiev, M. Shiozawa, Y. Shiraishi, A. Shvartsman, N. Skrobova, K.
Skwarczynski, D. Smyczek, M. Smy, J. T. Sobczyk, H. Sobel, F. J. P. Soler, A.
J. Speers, R. Spina, A. Srivastava, P. Stowell, Y. Stroke, I. A. Suslov, A.
Suzuki, S. Y. Suzuki, M. Tada, S. Tairafune, A. Takeda, Y. Takeuchi, K.
Takeya, H. K. Tanaka, H. Tanigawa, V. V. Tereshchenko, N. Thamm, C.
Touramanis, N. Tran, T. Tsukamoto, M. Tzanov, Y. Uchida, M. Vagins, S.
Valder, M. Varghese, I. Vasilyev, G. Vasseur, E. Villa, U. Virginet, T.
Vladisavljevic, T. Wachala, D. Wakabayashi, H. T. Wallace, J. G. Walsh, D.
Wark, M. O. Wascko, A. Weber, R. Wendell, M. J. Wilking, C. Wilkinson, J. R.
Wilson, K. Wood, C. Wret, J. Xia, K. Yamamoto, T. Yamamoto, C. Yanagisawa, Y.
Yang, T. Yano, N. Yershov, U. Yevarouskaya, M. Yokoyama, Y. Yoshimoto, N.
Yoshimura, R. Zaki, A. Zalewska, J. Zalipska, G. Zarnecki, J. Zhang, X. Y.
Zhao, H. Zheng, H. Zhong, T. Zhu, M. Ziembicki, E. D. Zimmerman, M. Zito, S.
Zsoldos | First Measurement of the Electron Neutrino Charged-Current Pion
Production Cross Section on Carbon with the T2K Near Detector | 8 pages, 2 figures. Data release: https://zenodo.org/records/15316318 | null | null | null | hep-ex | http://creativecommons.org/licenses/by/4.0/ | The T2K Collaboration presents the first measurement of electron
neutrino-induced charged-current pion production on carbon in a restricted
kinematical phase space. This is performed using data from the 2.5$^{\deg}$
off-axis near detector, ND280. The differential cross sections with respect to
the outgoing electron and pion kinematics, in addition to the total
flux-integrated cross section, are obtained. Comparisons between the measured
and predicted cross section results using the Neut, Genie and NuWro Monte Carlo
event generators are presented. The measured total flux-integrated cross
section is [2.52 $\pm$ 0.52 (stat) $\pm$ 0.30 (sys)] x $10^{-39}$ cm$^2$
nucleon$^{-1}$, which is lower than the event generator predictions.
| [
{
"version": "v1",
"created": "Thu, 1 May 2025 13:35:18 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Abe",
"K.",
""
],
[
"Abe",
"S.",
""
],
[
"Akutsu",
"R.",
""
],
[
"Alarakia-Charles",
"H.",
""
],
[
"Hakim",
"Y. I. Alj",
""
],
[
"Monsalve",
"S. Alonso",
""
],
[
"Anthony",
"L.",
""
],
[
"Aoki",
"S.",
""
],
[
"Apte",
"K. A.",
""
],
[
"Arai",
"T.",
""
],
[
"Arihara",
"T.",
""
],
[
"Arimoto",
"S.",
""
],
[
"Atkin",
"E. T.",
""
],
[
"Babu",
"N.",
""
],
[
"Baranov",
"V.",
""
],
[
"Barker",
"G. J.",
""
],
[
"Barr",
"G.",
""
],
[
"Barrow",
"D.",
""
],
[
"Bates",
"P.",
""
],
[
"Bathe-Peters",
"L.",
""
],
[
"Batkiewicz-Kwasniak",
"M.",
""
],
[
"Baudis",
"N.",
""
],
[
"Berardi",
"V.",
""
],
[
"Berns",
"L.",
""
],
[
"Bhattacharjee",
"S.",
""
],
[
"Blanchet",
"A.",
""
],
[
"Blondel",
"A.",
""
],
[
"Boistier",
"P. M. M.",
""
],
[
"Bolognesi",
"S.",
""
],
[
"Bordoni",
"S.",
""
],
[
"Boyd",
"S. B.",
""
],
[
"Bronner",
"C.",
""
],
[
"Bubak",
"A.",
""
],
[
"Avanzini",
"M. Buizza",
""
],
[
"Caballero",
"J. A.",
""
],
[
"Cadoux",
"F.",
""
],
[
"Calabria",
"N. F.",
""
],
[
"Cao",
"S.",
""
],
[
"Cap",
"S.",
""
],
[
"Carabadjac",
"D.",
""
],
[
"Cartwright",
"S. L.",
""
],
[
"Casado",
"M. P.",
""
],
[
"Catanesi",
"M. G.",
""
],
[
"Chakrani",
"J.",
""
],
[
"Chalumeau",
"A.",
""
],
[
"Cherdack",
"D.",
""
],
[
"Chvirova",
"A.",
""
],
[
"Coleman",
"J.",
""
],
[
"Collazuol",
"G.",
""
],
[
"Cormier",
"F.",
""
],
[
"Craplet",
"A. A. L.",
""
],
[
"Cudd",
"A.",
""
],
[
"D'ago",
"D.",
""
],
[
"Dalmazzone",
"C.",
""
],
[
"Daret",
"T.",
""
],
[
"Dasgupta",
"P.",
""
],
[
"Davis",
"C.",
""
],
[
"Davydov",
"Yu. I.",
""
],
[
"de Perio",
"P.",
""
],
[
"De Rosa",
"G.",
""
],
[
"Dealtry",
"T.",
""
],
[
"Densham",
"C.",
""
],
[
"Dergacheva",
"A.",
""
],
[
"Banerjee",
"R. Dharmapal",
""
],
[
"Di Lodovico",
"F.",
""
],
[
"Lopez",
"G. Diaz",
""
],
[
"Dolan",
"S.",
""
],
[
"Douqa",
"D.",
""
],
[
"Doyle",
"T. A.",
""
],
[
"Drapier",
"O.",
""
],
[
"Duffy",
"K. E.",
""
],
[
"Dumarchez",
"J.",
""
],
[
"Dunne",
"P.",
""
],
[
"Dygnarowicz",
"K.",
""
],
[
"Eguchi",
"A.",
""
],
[
"Elias",
"J.",
""
],
[
"Emery-Schrenk",
"S.",
""
],
[
"Erofeev",
"G.",
""
],
[
"Ershova",
"A.",
""
],
[
"Eurin",
"G.",
""
],
[
"Fedorova",
"D.",
""
],
[
"Fedotov",
"S.",
""
],
[
"Feltre",
"M.",
""
],
[
"Feng",
"L.",
""
],
[
"Ferlewicz",
"D.",
""
],
[
"Finch",
"A. J.",
""
],
[
"Fitton",
"M. D.",
""
],
[
"Forza",
"C.",
""
],
[
"Friend",
"M.",
""
],
[
"Fujii",
"Y.",
""
],
[
"Fukuda",
"Y.",
""
],
[
"Furui",
"Y.",
""
],
[
"García-Marcos",
"J.",
""
],
[
"Germer",
"A. C.",
""
],
[
"Giannessi",
"L.",
""
],
[
"Giganti",
"C.",
""
],
[
"Girgus",
"M.",
""
],
[
"Glagolev",
"V.",
""
],
[
"Gonin",
"M.",
""
],
[
"Jiménez",
"R. González",
""
],
[
"Rosa",
"J. González",
""
],
[
"Goodman",
"E. A. G.",
""
],
[
"Gorshanov",
"K.",
""
],
[
"Govindaraj",
"P.",
""
],
[
"Grassi",
"M.",
""
],
[
"Guigue",
"M.",
""
],
[
"Guo",
"F. Y.",
""
],
[
"Hadley",
"D. R.",
""
],
[
"Han",
"S.",
""
],
[
"Harris",
"D. A.",
""
],
[
"Harris",
"R. J.",
""
],
[
"Hasegawa",
"T.",
""
],
[
"Hasnip",
"C. M.",
""
],
[
"Hassani",
"S.",
""
],
[
"Hastings",
"N. C.",
""
],
[
"Hayato",
"Y.",
""
],
[
"Heitkamp",
"I.",
""
],
[
"Henaff",
"D.",
""
],
[
"Hino",
"Y.",
""
],
[
"Holeczek",
"J.",
""
],
[
"Holin",
"A.",
""
],
[
"Holvey",
"T.",
""
],
[
"Van",
"N. T. Hong",
""
],
[
"Honjo",
"T.",
""
],
[
"Hooft",
"M. C. F.",
""
],
[
"Hosokawa",
"K.",
""
],
[
"Hu",
"J.",
""
],
[
"Ichikawa",
"A. K.",
""
],
[
"Ieki",
"K.",
""
],
[
"Ikeda",
"M.",
""
],
[
"Ishida",
"T.",
""
],
[
"Ishitsuka",
"M.",
""
],
[
"Ito",
"H.",
""
],
[
"Ito",
"S.",
""
],
[
"Izmaylov",
"A.",
""
],
[
"Jachowicz",
"N.",
""
],
[
"Jenkins",
"S. J.",
""
],
[
"Jesús-Valls",
"C.",
""
],
[
"Jia",
"M.",
""
],
[
"Jiang",
"J. J.",
""
],
[
"Ji",
"J. Y.",
""
],
[
"Jones",
"T. P.",
""
],
[
"Jonsson",
"P.",
""
],
[
"Joshi",
"S.",
""
],
[
"Kabirnezhad",
"M.",
""
],
[
"Kaboth",
"A. C.",
""
],
[
"Kakuno",
"H.",
""
],
[
"Kameda",
"J.",
""
],
[
"Karpova",
"S.",
""
],
[
"Kasturi",
"V. S.",
""
],
[
"Kataoka",
"Y.",
""
],
[
"Katori",
"T.",
""
],
[
"Kawabata",
"A.",
""
],
[
"Kawamura",
"Y.",
""
],
[
"Kawaue",
"M.",
""
],
[
"Kearns",
"E.",
""
],
[
"Khabibullin",
"M.",
""
],
[
"Khotjantsev",
"A.",
""
],
[
"Kikawa",
"T.",
""
],
[
"King",
"S.",
""
],
[
"Kiseeva",
"V.",
""
],
[
"Kisiel",
"J.",
""
],
[
"Klustová",
"A.",
""
],
[
"Kneale",
"L.",
""
],
[
"Kobayashi",
"H.",
""
],
[
"Koch",
"L.",
""
],
[
"Kodama",
"S.",
""
],
[
"Kolupanova",
"M.",
""
],
[
"Konaka",
"A.",
""
],
[
"Kormos",
"L. L.",
""
],
[
"Koshio",
"Y.",
""
],
[
"Kowalik",
"K.",
""
],
[
"Kudenko",
"Y.",
""
],
[
"Kudo",
"Y.",
""
],
[
"Jha",
"A. Kumar",
""
],
[
"Kurjata",
"R.",
""
],
[
"Kurochka",
"V.",
""
],
[
"Kutter",
"T.",
""
],
[
"Labarga",
"L.",
""
],
[
"Lachat",
"M.",
""
],
[
"Lachner",
"K.",
""
],
[
"Lagoda",
"J.",
""
],
[
"Lakshmi",
"S. M.",
""
],
[
"James",
"M. Lamers",
""
],
[
"Langella",
"A.",
""
],
[
"Langridge",
"D. H.",
""
],
[
"Laporte",
"J. -F.",
""
],
[
"Last",
"D.",
""
],
[
"Latham",
"N.",
""
],
[
"Laveder",
"M.",
""
],
[
"Lavitola",
"L.",
""
],
[
"Lawe",
"M.",
""
],
[
"Silverio",
"D. Leon",
""
],
[
"Levorato",
"S.",
""
],
[
"Lewis",
"S. V.",
""
],
[
"Li",
"B.",
""
],
[
"Lin",
"C.",
""
],
[
"Litchfield",
"R. P.",
""
],
[
"Liu",
"S. L.",
""
],
[
"Li",
"W.",
""
],
[
"Longhin",
"A.",
""
],
[
"Moreno",
"A. Lopez",
""
],
[
"Ludovici",
"L.",
""
],
[
"Lu",
"X.",
""
],
[
"Lux",
"T.",
""
],
[
"Machado",
"L. N.",
""
],
[
"Magaletti",
"L.",
""
],
[
"Mahn",
"K.",
""
],
[
"Mahtani",
"K. K.",
""
],
[
"Manly",
"S.",
""
],
[
"Marino",
"A. D.",
""
],
[
"Martin",
"D. G. R.",
""
],
[
"Caicedo",
"D. A. Martinez",
""
],
[
"Martinez",
"L.",
""
],
[
"Martini",
"M.",
""
],
[
"Matsubara",
"T.",
""
],
[
"Matsumoto",
"R.",
""
],
[
"Matveev",
"V.",
""
],
[
"Mauger",
"C.",
""
],
[
"Mavrokoridis",
"K.",
""
],
[
"McCauley",
"N.",
""
],
[
"McFarland",
"K. S.",
""
],
[
"McGrew",
"C.",
""
],
[
"McKean",
"J.",
""
],
[
"Mefodiev",
"A.",
""
],
[
"Megias",
"G. D.",
""
],
[
"Mellet",
"L.",
""
],
[
"Metelko",
"C.",
""
],
[
"Mezzetto",
"M.",
""
],
[
"Miki",
"S.",
""
],
[
"Mikola",
"V.",
""
],
[
"Miller",
"E. W.",
""
],
[
"Minamino",
"A.",
""
],
[
"Mineev",
"O.",
""
],
[
"Mine",
"S.",
""
],
[
"Mirabito",
"J.",
""
],
[
"Miura",
"M.",
""
],
[
"Moriyama",
"S.",
""
],
[
"Moriyama",
"S.",
""
],
[
"Morrison",
"P.",
""
],
[
"Mueller",
"Th. A.",
""
],
[
"Munford",
"D.",
""
],
[
"Muñoz",
"A.",
""
],
[
"Munteanu",
"L.",
""
],
[
"Nagai",
"Y.",
""
],
[
"Nakadaira",
"T.",
""
],
[
"Nakagiri",
"K.",
""
],
[
"Nakahata",
"M.",
""
],
[
"Nakajima",
"Y.",
""
],
[
"Nakamura",
"K. D.",
""
],
[
"Nakano",
"A.",
""
],
[
"Nakano",
"Y.",
""
],
[
"Nakayama",
"S.",
""
],
[
"Nakaya",
"T.",
""
],
[
"Nakayoshi",
"K.",
""
],
[
"Naseby",
"C. E. R.",
""
],
[
"Nguyen",
"D. T.",
""
],
[
"Nguyen",
"V. Q.",
""
],
[
"Niewczas",
"K.",
""
],
[
"Nishimori",
"S.",
""
],
[
"Nishimura",
"Y.",
""
],
[
"Noguchi",
"Y.",
""
],
[
"Nosek",
"T.",
""
],
[
"Nova",
"F.",
""
],
[
"Nugent",
"J. C.",
""
],
[
"O'Keeffe",
"H. M.",
""
],
[
"O'Sullivan",
"L.",
""
],
[
"Okazaki",
"R.",
""
],
[
"Okinaga",
"W.",
""
],
[
"Okumura",
"K.",
""
],
[
"Okusawa",
"T.",
""
],
[
"Onda",
"N.",
""
],
[
"Ospina",
"N.",
""
],
[
"Osu",
"L.",
""
],
[
"Oyama",
"Y.",
""
],
[
"Paolone",
"V.",
""
],
[
"Pasternak",
"J.",
""
],
[
"Payne",
"D.",
""
],
[
"Pfaff",
"M.",
""
],
[
"Pickering",
"L.",
""
],
[
"Popov",
"B.",
""
],
[
"Yrey",
"A. J. Portocarrero",
""
],
[
"Posiadala-Zezula",
"M.",
""
],
[
"Prabhu",
"Y. S.",
""
],
[
"Prasad",
"H.",
""
],
[
"Pupilli",
"F.",
""
],
[
"Quilain",
"B.",
""
],
[
"Quyen",
"P. T.",
""
],
[
"Radicioni",
"E.",
""
],
[
"Radics",
"B.",
""
],
[
"Ramirez",
"M. A.",
""
],
[
"Ramsden",
"R.",
""
],
[
"Ratoff",
"P. N.",
""
],
[
"Reh",
"M.",
""
],
[
"Reina",
"G.",
""
],
[
"Riccio",
"C.",
""
],
[
"Riley",
"D. W.",
""
],
[
"Rondio",
"E.",
""
],
[
"Roth",
"S.",
""
],
[
"Roy",
"N.",
""
],
[
"Rubbia",
"A.",
""
],
[
"Russo",
"L.",
""
],
[
"Rychter",
"A.",
""
],
[
"Saenz",
"W.",
""
],
[
"Sakashita",
"K.",
""
],
[
"Samani",
"S.",
""
],
[
"Sánchez",
"F.",
""
],
[
"Sandford",
"E. M.",
""
],
[
"Sato",
"Y.",
""
],
[
"Schefke",
"T.",
""
],
[
"Scholberg",
"K.",
""
],
[
"Scott",
"M.",
""
],
[
"Seiya",
"Y.",
""
],
[
"Sekiguchi",
"T.",
""
],
[
"Sekiya",
"H.",
""
],
[
"Sekiya",
"T.",
""
],
[
"Seppala",
"D.",
""
],
[
"Sgalaberna",
"D.",
""
],
[
"Shaikhiev",
"A.",
""
],
[
"Shiozawa",
"M.",
""
],
[
"Shiraishi",
"Y.",
""
],
[
"Shvartsman",
"A.",
""
],
[
"Skrobova",
"N.",
""
],
[
"Skwarczynski",
"K.",
""
],
[
"Smyczek",
"D.",
""
],
[
"Smy",
"M.",
""
],
[
"Sobczyk",
"J. T.",
""
],
[
"Sobel",
"H.",
""
],
[
"Soler",
"F. J. P.",
""
],
[
"Speers",
"A. J.",
""
],
[
"Spina",
"R.",
""
],
[
"Srivastava",
"A.",
""
],
[
"Stowell",
"P.",
""
],
[
"Stroke",
"Y.",
""
],
[
"Suslov",
"I. A.",
""
],
[
"Suzuki",
"A.",
""
],
[
"Suzuki",
"S. Y.",
""
],
[
"Tada",
"M.",
""
],
[
"Tairafune",
"S.",
""
],
[
"Takeda",
"A.",
""
],
[
"Takeuchi",
"Y.",
""
],
[
"Takeya",
"K.",
""
],
[
"Tanaka",
"H. K.",
""
],
[
"Tanigawa",
"H.",
""
],
[
"Tereshchenko",
"V. V.",
""
],
[
"Thamm",
"N.",
""
],
[
"Touramanis",
"C.",
""
],
[
"Tran",
"N.",
""
],
[
"Tsukamoto",
"T.",
""
],
[
"Tzanov",
"M.",
""
],
[
"Uchida",
"Y.",
""
],
[
"Vagins",
"M.",
""
],
[
"Valder",
"S.",
""
],
[
"Varghese",
"M.",
""
],
[
"Vasilyev",
"I.",
""
],
[
"Vasseur",
"G.",
""
],
[
"Villa",
"E.",
""
],
[
"Virginet",
"U.",
""
],
[
"Vladisavljevic",
"T.",
""
],
[
"Wachala",
"T.",
""
],
[
"Wakabayashi",
"D.",
""
],
[
"Wallace",
"H. T.",
""
],
[
"Walsh",
"J. G.",
""
],
[
"Wark",
"D.",
""
],
[
"Wascko",
"M. O.",
""
],
[
"Weber",
"A.",
""
],
[
"Wendell",
"R.",
""
],
[
"Wilking",
"M. J.",
""
],
[
"Wilkinson",
"C.",
""
],
[
"Wilson",
"J. R.",
""
],
[
"Wood",
"K.",
""
],
[
"Wret",
"C.",
""
],
[
"Xia",
"J.",
""
],
[
"Yamamoto",
"K.",
""
],
[
"Yamamoto",
"T.",
""
],
[
"Yanagisawa",
"C.",
""
],
[
"Yang",
"Y.",
""
],
[
"Yano",
"T.",
""
],
[
"Yershov",
"N.",
""
],
[
"Yevarouskaya",
"U.",
""
],
[
"Yokoyama",
"M.",
""
],
[
"Yoshimoto",
"Y.",
""
],
[
"Yoshimura",
"N.",
""
],
[
"Zaki",
"R.",
""
],
[
"Zalewska",
"A.",
""
],
[
"Zalipska",
"J.",
""
],
[
"Zarnecki",
"G.",
""
],
[
"Zhang",
"J.",
""
],
[
"Zhao",
"X. Y.",
""
],
[
"Zheng",
"H.",
""
],
[
"Zhong",
"H.",
""
],
[
"Zhu",
"T.",
""
],
[
"Ziembicki",
"M.",
""
],
[
"Zimmerman",
"E. D.",
""
],
[
"Zito",
"M.",
""
],
[
"Zsoldos",
"S.",
""
]
] |
2505.00536 | Devon Lin Professor | Guangzhou Chen, Yuanzhen He, C. Devon Lin and Fasheng Sun | Grouped Orthogonal Arrays and Their Applications | null | Statistica Sinica, 2025 | null | null | stat.ME | http://creativecommons.org/licenses/by/4.0/ | In computer experiments, it has become a standard practice to select the
inputs that spread out as uniformly as possible over the design space. The
resulting designs are called space-filling designs and they are undoubtedly
desirable choices when there is no prior knowledge on how the input variables
affect the response and the objective of experiments is global fitting. When
there is some prior knowledge on the underlying true function of the system or
what statistical models are more appropriate, a natural question is, are there
more suitable designs than vanilla space-filling designs? In this article, we
provide an answer for the cases where there are no interactions between the
factors from disjoint groups of variables. In other words, we consider the
design issue when the underlying functional form of the system or the
statistical model to be used is additive where each component depends on one
group of variables from a set of disjoint groups. For such cases, we recommend
using {\em grouped orthogonal arrays.} Several construction methods are
provided and many designs are tabulated for practical use. Compared with
existing techniques in the literature, our construction methods can generate
many more designs with flexible run sizes and better within-group projection
properties for any prime power number of levels.
| [
{
"version": "v1",
"created": "Thu, 1 May 2025 14:02:35 GMT"
},
{
"version": "v2",
"created": "Mon, 5 May 2025 14:12:15 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Chen",
"Guangzhou",
""
],
[
"He",
"Yuanzhen",
""
],
[
"Lin",
"C. Devon",
""
],
[
"Sun",
"Fasheng",
""
]
] |
2505.00578 | Shuang Zhang | Shuang Zhang, Carleton Coffin, Karyn L. Rogers, Catherine Ann Royer,
Ge Wang | AI-Driven Segmentation and Analysis of Microbial Cells | null | null | null | null | eess.IV q-bio.QM | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Studying the growth and metabolism of microbes provides critical insights
into their evolutionary adaptations to harsh environments, which are essential
for microbial research and biotechnology applications. In this study, we
developed an AI-driven image analysis system to efficiently segment individual
cells and quantitatively analyze key cellular features. This system is
comprised of four main modules. First, a denoising algorithm enhances contrast
and suppresses noise while preserving fine cellular details. Second, the
Segment Anything Model (SAM) enables accurate, zero-shot segmentation of cells
without additional training. Third, post-processing is applied to refine
segmentation results by removing over-segmented masks. Finally, quantitative
analysis algorithms extract essential cellular features, including average
intensity, length, width, and volume. The results show that denoising and
post-processing significantly improved the segmentation accuracy of SAM in this
new domain. Without human annotations, the AI-driven pipeline automatically and
efficiently outlines cellular boundaries, indexes them, and calculates key
cellular parameters with high accuracy. This framework will enable efficient
and automated quantitative analysis of high-resolution fluorescence microscopy
images to advance research into microbial adaptations to grow and metabolism
that allow extremophiles to thrive in their harsh habitats.
| [
{
"version": "v1",
"created": "Thu, 1 May 2025 15:09:03 GMT"
},
{
"version": "v2",
"created": "Mon, 5 May 2025 16:24:34 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Zhang",
"Shuang",
""
],
[
"Coffin",
"Carleton",
""
],
[
"Rogers",
"Karyn L.",
""
],
[
"Royer",
"Catherine Ann",
""
],
[
"Wang",
"Ge",
""
]
] |
2505.00626 | Zihao Wang | Zihao Wang, Yibo Jiang, Jiahao Yu, Heqing Huang | The Illusion of Role Separation: Hidden Shortcuts in LLM Role Learning
(and How to Fix Them) | null | null | null | null | cs.CL cs.AI | http://creativecommons.org/licenses/by/4.0/ | Large language models (LLMs) that integrate multiple input roles (e.g.,
system instructions, user queries, external tool outputs) are increasingly
prevalent in practice. Ensuring that the model accurately distinguishes
messages from each role -- a concept we call \emph{role separation} -- is
crucial for consistent multi-role behavior. Although recent work often targets
state-of-the-art prompt injection defenses, it remains unclear whether such
methods truly teach LLMs to differentiate roles or merely memorize known
triggers. In this paper, we examine \emph{role-separation learning}: the
process of teaching LLMs to robustly distinguish system and user tokens.
Through a \emph{simple, controlled experimental framework}, we find that
fine-tuned models often rely on two proxies for role identification: (1) task
type exploitation, and (2) proximity to begin-of-text. Although data
augmentation can partially mitigate these shortcuts, it generally leads to
iterative patching rather than a deeper fix. To address this, we propose
reinforcing \emph{invariant signals} that mark role boundaries by adjusting
token-wise cues in the model's input encoding. In particular, manipulating
position IDs helps the model learn clearer distinctions and reduces reliance on
superficial proxies. By focusing on this mechanism-centered perspective, our
work illuminates how LLMs can more reliably maintain consistent multi-role
behavior without merely memorizing known prompts or triggers.
| [
{
"version": "v1",
"created": "Thu, 1 May 2025 16:06:16 GMT"
},
{
"version": "v2",
"created": "Mon, 5 May 2025 03:29:08 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Wang",
"Zihao",
""
],
[
"Jiang",
"Yibo",
""
],
[
"Yu",
"Jiahao",
""
],
[
"Huang",
"Heqing",
""
]
] |
2505.00630 | Muyi Bao | Muyi Bao, Shuchang Lyu, Zhaoyang Xu, Huiyu Zhou, Jinchang Ren, Shiming
Xiang, Xiangtai Li, Guangliang Cheng | Vision Mamba in Remote Sensing: A Comprehensive Survey of Techniques,
Applications and Outlook | null | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Deep learning has profoundly transformed remote sensing, yet prevailing
architectures like Convolutional Neural Networks (CNNs) and Vision Transformers
(ViTs) remain constrained by critical trade-offs: CNNs suffer from limited
receptive fields, while ViTs grapple with quadratic computational complexity,
hindering their scalability for high-resolution remote sensing data. State
Space Models (SSMs), particularly the recently proposed Mamba architecture,
have emerged as a paradigm-shifting solution, combining linear computational
scaling with global context modeling. This survey presents a comprehensive
review of Mamba-based methodologies in remote sensing, systematically analyzing
about 120 Mamba-based remote sensing studies to construct a holistic taxonomy
of innovations and applications. Our contributions are structured across five
dimensions: (i) foundational principles of vision Mamba architectures, (ii)
micro-architectural advancements such as adaptive scan strategies and hybrid
SSM formulations, (iii) macro-architectural integrations, including
CNN-Transformer-Mamba hybrids and frequency-domain adaptations, (iv) rigorous
benchmarking against state-of-the-art methods in multiple application tasks,
such as object detection, semantic segmentation, change detection, etc. and (v)
critical analysis of unresolved challenges with actionable future directions.
By bridging the gap between SSM theory and remote sensing practice, this survey
establishes Mamba as a transformative framework for remote sensing analysis. To
our knowledge, this paper is the first systematic review of Mamba architectures
in remote sensing. Our work provides a structured foundation for advancing
research in remote sensing systems through SSM-based methods. We curate an
open-source repository
(https://github.com/BaoBao0926/Awesome-Mamba-in-Remote-Sensing) to foster
community-driven advancements.
| [
{
"version": "v1",
"created": "Thu, 1 May 2025 16:07:51 GMT"
},
{
"version": "v2",
"created": "Sat, 3 May 2025 09:38:12 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Bao",
"Muyi",
""
],
[
"Lyu",
"Shuchang",
""
],
[
"Xu",
"Zhaoyang",
""
],
[
"Zhou",
"Huiyu",
""
],
[
"Ren",
"Jinchang",
""
],
[
"Xiang",
"Shiming",
""
],
[
"Li",
"Xiangtai",
""
],
[
"Cheng",
"Guangliang",
""
]
] |
2505.00641 | Nan An | Nan An | Expected First Return Times for Random Walks on Bounded Grids | null | null | null | null | math.PR | http://creativecommons.org/licenses/by/4.0/ | We derive a general formula for computing the expected first return time of a
random walk on a finite graph. Using this framework, we calculate the expected
first return time in various settings over bounded rectangular grids with
different boundary conditions.
| [
{
"version": "v1",
"created": "Thu, 1 May 2025 16:30:55 GMT"
},
{
"version": "v2",
"created": "Fri, 2 May 2025 21:56:21 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"An",
"Nan",
""
]
] |
2505.00722 | Tarapada Bag | Abhishikta Das, Hemanta Kalita, Mohammad Sajid, T. Bag | Generalized $\theta$-Parametric Metric Spaces: Fixed Point Theorems and
Applications to Fractional Economic Models | 23 pages | null | null | null | math.OC | http://creativecommons.org/licenses/by-nc-nd/4.0/ | The objective of this manuscript is to introduce and develop the concept of a
generalized $\theta$-parametric metric space-a novel extension that enriches
the modern metric fixed point theory. We study of its fundamental properties,
including convergence and Cauchy sequences that establishes a solid theoretical
foundation. A significant highlight of our work is the formulation of
Suzuki-type fixed point theorem within this framework which extends classical
results in a meaningful way. To demonstrate the depth and applicability of our
findings, we construct non-trivial examples that illustrate the behavior of key
concepts. Moreover, as a practical application, we apply our main theorem to
analyze an economic growth model, demonstrating its utility in solving
fractional differential equations that arise in dynamic economic systems.
| [
{
"version": "v1",
"created": "Tue, 22 Apr 2025 08:16:23 GMT"
},
{
"version": "v2",
"created": "Mon, 5 May 2025 09:51:34 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Das",
"Abhishikta",
""
],
[
"Kalita",
"Hemanta",
""
],
[
"Sajid",
"Mohammad",
""
],
[
"Bag",
"T.",
""
]
] |
2505.00744 | Vu Minh Hieu Phan Dr | Dung Nguyen, Minh Khoi Ho, Huy Ta, Thanh Tam Nguyen, Qi Chen, Kumar
Rav, Quy Duong Dang, Satwik Ramchandre, Son Lam Phung, Zhibin Liao, Minh-Son
To, Johan Verjans, Phi Le Nguyen, and Vu Minh Hieu Phan | Localizing Before Answering: A Hallucination Evaluation Benchmark for
Grounded Medical Multimodal LLMs | Accepted at Joint Conference on Artificial Intelligence (IJCAI) 2025 | null | null | null | cs.CV | http://creativecommons.org/publicdomain/zero/1.0/ | Medical Large Multi-modal Models (LMMs) have demonstrated remarkable
capabilities in medical data interpretation. However, these models frequently
generate hallucinations contradicting source evidence, particularly due to
inadequate localization reasoning. This work reveals a critical limitation in
current medical LMMs: instead of analyzing relevant pathological regions, they
often rely on linguistic patterns or attend to irrelevant image areas when
responding to disease-related queries. To address this, we introduce
HEAL-MedVQA (Hallucination Evaluation via Localization MedVQA), a comprehensive
benchmark designed to evaluate LMMs' localization abilities and hallucination
robustness. HEAL-MedVQA features (i) two innovative evaluation protocols to
assess visual and textual shortcut learning, and (ii) a dataset of 67K VQA
pairs, with doctor-annotated anatomical segmentation masks for pathological
regions. To improve visual reasoning, we propose the Localize-before-Answer
(LobA) framework, which trains LMMs to localize target regions of interest and
self-prompt to emphasize segmented pathological areas, generating grounded and
reliable answers. Experimental results demonstrate that our approach
significantly outperforms state-of-the-art biomedical LMMs on the challenging
HEAL-MedVQA benchmark, advancing robustness in medical VQA.
| [
{
"version": "v1",
"created": "Wed, 30 Apr 2025 07:57:51 GMT"
},
{
"version": "v2",
"created": "Mon, 5 May 2025 02:30:17 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Nguyen",
"Dung",
""
],
[
"Ho",
"Minh Khoi",
""
],
[
"Ta",
"Huy",
""
],
[
"Nguyen",
"Thanh Tam",
""
],
[
"Chen",
"Qi",
""
],
[
"Rav",
"Kumar",
""
],
[
"Dang",
"Quy Duong",
""
],
[
"Ramchandre",
"Satwik",
""
],
[
"Phung",
"Son Lam",
""
],
[
"Liao",
"Zhibin",
""
],
[
"To",
"Minh-Son",
""
],
[
"Verjans",
"Johan",
""
],
[
"Nguyen",
"Phi Le",
""
],
[
"Phan",
"Vu Minh Hieu",
""
]
] |
2505.00985 | Ayan Sengupta | Ayan Sengupta, Yash Goel, Tanmoy Chakraborty | Position: Enough of Scaling LLMs! Lets Focus on Downscaling | null | null | null | null | cs.CL | http://creativecommons.org/licenses/by/4.0/ | We challenge the dominant focus on neural scaling laws and advocate for a
paradigm shift toward downscaling in the development of large language models
(LLMs). While scaling laws have provided critical insights into performance
improvements through increasing model and dataset size, we emphasize the
significant limitations of this approach, particularly in terms of
computational inefficiency, environmental impact, and deployment constraints.
To address these challenges, we propose a holistic framework for downscaling
LLMs that seeks to maintain performance while drastically reducing resource
demands. This paper outlines practical strategies for transitioning away from
traditional scaling paradigms, advocating for a more sustainable, efficient,
and accessible approach to LLM development.
| [
{
"version": "v1",
"created": "Fri, 2 May 2025 04:13:27 GMT"
},
{
"version": "v2",
"created": "Mon, 5 May 2025 04:30:12 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Sengupta",
"Ayan",
""
],
[
"Goel",
"Yash",
""
],
[
"Chakraborty",
"Tanmoy",
""
]
] |
2505.01066 | Karoly J. Boroczky | Karoly J. Boroczky, Shibing Chen, Weiru Liu, Christos Saroglou | Uniqueness in the near isotropic Lp dual Minkowski problem | null | null | null | null | math.AP | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | For n>1 and -1<p<1, we prove that if q is close to n and the qth Lp dual
curvature is Holder close to be the constant one function, then this "near
isotropic" qth Lp dual Minkowski problem on the (n-1)-dimensional sphere has a
unique solution. Along the way, we establish a C0 estimate for -1<p<1 that is
optimal in the sense that if p<-1 and q=n, then it is known that the analogous
C0 estimate does not hold. We also prove the uniqueness of the solution of the
near isotropic even qth Lp dual Minkowski problem on the (n-1)-dimensional
sphere if -1<p<q<min{n,n+p} and q>0.
| [
{
"version": "v1",
"created": "Fri, 2 May 2025 07:15:33 GMT"
},
{
"version": "v2",
"created": "Mon, 5 May 2025 12:49:01 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Boroczky",
"Karoly J.",
""
],
[
"Chen",
"Shibing",
""
],
[
"Liu",
"Weiru",
""
],
[
"Saroglou",
"Christos",
""
]
] |
2505.01084 | Kentaro Inoue | Kentaro Inoue | Log prismatic $F$-crystals and realization functors | 69 pages. Some contents of this article previously belonged to
arXiv:2503.07379 | null | null | null | math.AG math.NT | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Log prismatic cohomology theory developed by Koshikawa-Yao involves
coefficient objects, called log prismatic $F$-crystals. In this paper, we
construct and study realization functors from the category of log prismatic
$F$-crystals to the category of coefficient objects of other $p$-adic
cohomology theories, in the setting where boundary divisors may involve
horizontal components.
| [
{
"version": "v1",
"created": "Fri, 2 May 2025 07:45:30 GMT"
},
{
"version": "v2",
"created": "Mon, 5 May 2025 05:54:20 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Inoue",
"Kentaro",
""
]
] |
2505.01138 | Matteo Casati | Guido Carlet, Matteo Casati | On the structure of homogeneous local Poisson brackets | 18 pages | null | null | null | math.DG math-ph math.MP nlin.SI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We consider an arbitrary Dubrovin-Novikov bracket of degree $k$, namely a
homogeneous degree $k$ local Poisson bracket on the loop space of a smooth
manifold $M$ of dimension $n$, and show that $k$ connections, defined by
explicit linear combinations with constant coefficients of the standard
connections associated with the Poisson bracket, are flat.
| [
{
"version": "v1",
"created": "Fri, 2 May 2025 09:26:31 GMT"
},
{
"version": "v2",
"created": "Mon, 5 May 2025 09:47:23 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Carlet",
"Guido",
""
],
[
"Casati",
"Matteo",
""
]
] |
2505.01141 | Viktor Mandrolko | Viktor Mandrolko, Konstantinos Termentzidis, David Lacroix and Mykola
Isaiev | Tailoring Heat Transfer at Silica-Water Interfaces via Hydroxyl and
Methyl Surface Groups | null | null | null | null | physics.atm-clus cond-mat.mtrl-sci | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Efficient thermal transport across solid-liquid interfaces is essential for
optimizing heat dissipation in modern technological applications. This study
employs molecular dynamics (MD) simulations to investigate the impact of
surface functionalization on heat transfer at the silica/water interface. It
has been shown that the surface functionalization changes significantly the
wetting characteristics of silica surface: from one hand hydroxyl groups render
such surfaces more hydrophilic, while methyl groups more hydrophobic. Here, we
reveal that modifying the surface functionalization from methylated to
hydroxylated groups results in: (i) up to an approximately eightfold increase
in adhesion energy, (ii) a reorientation of interfacial water molecules to
align perpendicular to the surface normal, (iii) a reduction in the liquid
depletion length near the interface, and (iv) an overall enhancement of
interfacial heat conduction. We quantify interfacial thermal resistance through
the calculation of the contribution of each functional group to the total heat
flux, providing insights into the physical mechanisms governing heat transfer
at functionalized interfaces. We demonstrated that manipulation of the
concentrations of the functional groups can be used to tailor interfacial
thermal transport.
| [
{
"version": "v1",
"created": "Fri, 2 May 2025 09:34:19 GMT"
},
{
"version": "v2",
"created": "Mon, 5 May 2025 06:42:49 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Mandrolko",
"Viktor",
""
],
[
"Termentzidis",
"Konstantinos",
""
],
[
"Lacroix",
"David",
""
],
[
"Isaiev",
"Mykola",
""
]
] |
2505.01152 | Tianfu Qi | Tianfu Qi, Jun Wang | Polarization Decomposition and Its Applications | null | null | null | null | cs.IT math.IT | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The polarization decomposition of arbitrary binary-input memoryless channels
(BMCs) is studied in this work. By introducing the polarization factor (PF),
defined in terms of the conditional entropy of the channel output under various
input configurations, we demonstrate that the symmetric capacities of the
polarized subchannels can be uniformly expressed as functions of the PF. The
explicit formulation of the PF as a function of the block length and subchannel
index is derived. Furthermore, an efficient algorithm is proposed for the
computation of the PF. Notably, we establish a one-to-one correspondence
between each PF and an $n$-ary tree. Leveraging this tree structure, we develop
a pruning method to determine the conditional entropy associated with different
input relationships. The proposed polarization framework offers both
theoretical insights and practical advantages, including intuitive
visualization of polarization behavior and efficient polar code construction.
To the best of our knowledge, this is the first approach that enables the
efficient computation of symmetric capacities for all subchannels in arbitrary
BMCs.
| [
{
"version": "v1",
"created": "Fri, 2 May 2025 09:53:31 GMT"
},
{
"version": "v2",
"created": "Mon, 5 May 2025 09:25:05 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Qi",
"Tianfu",
""
],
[
"Wang",
"Jun",
""
]
] |
2505.01167 | Gabriel Ramos-Fernandez | Gabriel Ramos-Fernandez, Ross S. Walker, Matthew J. Silk, Denis Boyer
and Sandra E. Smith-Aguilar | Uncovering complementary information sharing in spider monkey collective
foraging using higher-order spatial networks | 18 pages, 6 figures plus supplementary information with 12 pages,
methods, model and 4 figures of results | null | null | null | q-bio.QM cond-mat.dis-nn cond-mat.stat-mech cs.SI physics.soc-ph | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Collectives are often able to process information in a distributed fashion,
surpassing each individual member's processing capacity. In fission-fusion
dynamics, where group members come together and split from others often,
sharing complementary information about uniquely known foraging areas could
allow a group to track a heterogenous foraging environment better than any
group member on its own. We analyse the partial overlaps between individual
core ranges, which we assume represent the knowledge of an individual during a
given season. We identify sets of individuals whose overlap shows a balance
between redundantly and uniquely known portions and we use simplicial complexes
to represent these higher-order interactions. The structure of the simplicial
complexes shows holes in various dimensions, revealing complementarity in the
foraging information that is being shared. We propose that the complex spatial
networks arising from fission-fusion dynamics allow for adaptive, collective
processing of foraging information in dynamic environments.
| [
{
"version": "v1",
"created": "Fri, 2 May 2025 10:16:58 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Ramos-Fernandez",
"Gabriel",
""
],
[
"Walker",
"Ross S.",
""
],
[
"Silk",
"Matthew J.",
""
],
[
"Boyer",
"Denis",
""
],
[
"Smith-Aguilar",
"Sandra E.",
""
]
] |
2505.01182 | Ziyan Guo | Ziyan Guo and Haoxuan Qu and Hossein Rahmani and Dewen Soh and Ping Hu
and Qiuhong Ke and Jun Liu | TSTMotion: Training-free Scene-aware Text-to-motion Generation | Accepted by ICME2025 | null | null | null | cs.CV cs.AI | http://creativecommons.org/licenses/by/4.0/ | Text-to-motion generation has recently garnered significant research
interest, primarily focusing on generating human motion sequences in blank
backgrounds. However, human motions commonly occur within diverse 3D scenes,
which has prompted exploration into scene-aware text-to-motion generation
methods. Yet, existing scene-aware methods often rely on large-scale
ground-truth motion sequences in diverse 3D scenes, which poses practical
challenges due to the expensive cost. To mitigate this challenge, we are the
first to propose a \textbf{T}raining-free \textbf{S}cene-aware
\textbf{T}ext-to-\textbf{Motion} framework, dubbed as \textbf{TSTMotion}, that
efficiently empowers pre-trained blank-background motion generators with the
scene-aware capability. Specifically, conditioned on the given 3D scene and
text description, we adopt foundation models together to reason, predict and
validate a scene-aware motion guidance. Then, the motion guidance is
incorporated into the blank-background motion generators with two
modifications, resulting in scene-aware text-driven motion sequences. Extensive
experiments demonstrate the efficacy and generalizability of our proposed
framework. We release our code in \href{https://tstmotion.github.io/}{Project
Page}.
| [
{
"version": "v1",
"created": "Fri, 2 May 2025 10:50:04 GMT"
},
{
"version": "v2",
"created": "Mon, 5 May 2025 05:14:20 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Guo",
"Ziyan",
""
],
[
"Qu",
"Haoxuan",
""
],
[
"Rahmani",
"Hossein",
""
],
[
"Soh",
"Dewen",
""
],
[
"Hu",
"Ping",
""
],
[
"Ke",
"Qiuhong",
""
],
[
"Liu",
"Jun",
""
]
] |
2505.01184 | Mar Tejedor | Mar Tejedor, Berta Casas, Javier Conejero, Alba Cervera-Lierta, Rosa
M. Badia | Distributed Quantum Circuit Cutting for Hybrid Quantum-Classical
High-Performance Computing | 12 pages, 13 figures, 1 table | null | null | null | cs.DC quant-ph | http://creativecommons.org/licenses/by/4.0/ | Most quantum computers today are constrained by hardware limitations,
particularly the number of available qubits, causing significant challenges for
executing large-scale quantum algorithms. Circuit cutting has emerged as a key
technique to overcome these limitations by decomposing large quantum circuits
into smaller subcircuits that can be executed independently and later
reconstructed. In this work, we introduce Qdislib, a distributed and flexible
library for quantum circuit cutting, designed to seamlessly integrate with
hybrid quantum-classical high-performance computing (HPC) systems. Qdislib
employs a graph-based representation of quantum circuits to enable efficient
partitioning, manipulation and execution, supporting both wire cutting and gate
cutting techniques. The library is compatible with multiple quantum computing
libraries, including Qiskit and Qibo, and leverages distributed computing
frameworks to execute subcircuits across CPUs, GPUs, and quantum processing
units (QPUs) in a fully parallelized manner. We present a proof of concept
demonstrating how Qdislib enables the distributed execution of quantum circuits
across heterogeneous computing resources, showcasing its potential for scalable
quantum-classical workflows.
| [
{
"version": "v1",
"created": "Fri, 2 May 2025 11:00:34 GMT"
},
{
"version": "v2",
"created": "Mon, 5 May 2025 08:43:29 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Tejedor",
"Mar",
""
],
[
"Casas",
"Berta",
""
],
[
"Conejero",
"Javier",
""
],
[
"Cervera-Lierta",
"Alba",
""
],
[
"Badia",
"Rosa M.",
""
]
] |
2505.01209 | Shunpu Tang | Shunpu Tang, Yuanyuan Jia, Qianqian Yang, Ruichen Zhang, Jihong Park,
Dusit Niyato | Enabling Training-Free Semantic Communication Systems with Generative
Diffusion Models | null | null | null | null | cs.IT eess.SP math.IT | http://creativecommons.org/licenses/by/4.0/ | Semantic communication (SemCom) has recently emerged as a promising paradigm
for next-generation wireless systems. Empowered by advanced artificial
intelligence (AI) technologies, SemCom has achieved significant improvements in
transmission quality and efficiency. However, existing SemCom systems either
rely on training over large datasets and specific channel conditions or suffer
from performance degradation under channel noise when operating in a
training-free manner. To address these issues, we explore the use of generative
diffusion models (GDMs) as training-free SemCom systems. Specifically, we
design a semantic encoding and decoding method based on the inversion and
sampling process of the denoising diffusion implicit model (DDIM), which
introduces a two-stage forward diffusion process, split between the transmitter
and receiver to enhance robustness against channel noise. Moreover, we optimize
sampling steps to compensate for the increased noise level caused by channel
noise. We also conduct a brief analysis to provide insights about this design.
Simulations on the Kodak dataset validate that the proposed system outperforms
the existing baseline SemCom systems across various metrics.
| [
{
"version": "v1",
"created": "Fri, 2 May 2025 11:53:45 GMT"
},
{
"version": "v2",
"created": "Mon, 5 May 2025 02:49:04 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Tang",
"Shunpu",
""
],
[
"Jia",
"Yuanyuan",
""
],
[
"Yang",
"Qianqian",
""
],
[
"Zhang",
"Ruichen",
""
],
[
"Park",
"Jihong",
""
],
[
"Niyato",
"Dusit",
""
]
] |
2505.01251 | Galo Paez Fajardo | Galo J. Paez Fajardo, Daniela Dogaru, Muhammad Ans, Matthew J. W.
Ogley, Veronika Majherova, Innes McClelland, Shohei Hayashida, Pascal Puphal,
Masahiko Isobe, Bernhard Keimer, Pardeep K. Thakur, Tien-Lin Lee, Dave C.
Grinter, Pilar Ferrer, Serena A. Cussen, Matthias Hepting, Louis F. J. Piper | Direct Evidence of Metal-Ligand Redox in Li-ion Battery Cathodes | null | null | null | null | cond-mat.mtrl-sci | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Describing Li-ion battery cathodes in terms of distinct transition metal or
oxygen redox regimes can lead to confusion in understanding metal-ligand
hybridisation, oxygen dimerisation, and degradation. There is a pressing need
to study the electronic structure of these materials and determine the role
each cation and anion plays in charge compensation. Here, we employ transition
metal Ledge X-ray Resonance Photoemission Spectroscopy in conjunction with
Single Impurity Anderson simulations to directly evaluate the redox mechanisms
in (de-)lithiated battery electrodes. This approach reconciles the redox
description of two canonical cathodes -- LiMn$_{0.6}$Fe$_{0.4}$PO$_{4}$ and
LiNiO$_{2}$ -- in terms of varying degrees of charge transfer using the
established Zaanen-Sawatzky-Allen framework, common to condensed matter
physics. In LiMn$_{0.6}$Fe$_{0.4}$PO$_{4}$, we show that capacity arises due to
the depopulation of metal 3d states. Whereas, in LiNiO$_{2}$, charge transfer
dominates and redox occurs through the formation and elimination of ligand hole
states. This work clarifies the role of oxygen in Ni-rich system and provides a
framework to explain how capacity can be extracted from oxygen-dominated states
in highly covalent systems without needing to invoke dimerisation.
| [
{
"version": "v1",
"created": "Fri, 2 May 2025 13:18:59 GMT"
},
{
"version": "v2",
"created": "Mon, 5 May 2025 06:07:28 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Fajardo",
"Galo J. Paez",
""
],
[
"Dogaru",
"Daniela",
""
],
[
"Ans",
"Muhammad",
""
],
[
"Ogley",
"Matthew J. W.",
""
],
[
"Majherova",
"Veronika",
""
],
[
"McClelland",
"Innes",
""
],
[
"Hayashida",
"Shohei",
""
],
[
"Puphal",
"Pascal",
""
],
[
"Isobe",
"Masahiko",
""
],
[
"Keimer",
"Bernhard",
""
],
[
"Thakur",
"Pardeep K.",
""
],
[
"Lee",
"Tien-Lin",
""
],
[
"Grinter",
"Dave C.",
""
],
[
"Ferrer",
"Pilar",
""
],
[
"Cussen",
"Serena A.",
""
],
[
"Hepting",
"Matthias",
""
],
[
"Piper",
"Louis F. J.",
""
]
] |
2505.01315 | Sheikh Samit Muhaimin | Sheikh Samit Muhaimin, Spyridon Mastorakis | Helping Large Language Models Protect Themselves: An Enhanced Filtering
and Summarization System | null | null | null | null | cs.CL cs.AI | http://creativecommons.org/licenses/by/4.0/ | The recent growth in the use of Large Language Models has made them
vulnerable to sophisticated adversarial assaults, manipulative prompts, and
encoded malicious inputs. Existing countermeasures frequently necessitate
retraining models, which is computationally costly and impracticable for
deployment. Without the need for retraining or fine-tuning, this study presents
a unique defense paradigm that allows LLMs to recognize, filter, and defend
against adversarial or malicious inputs on their own. There are two main parts
to the suggested framework: (1) A prompt filtering module that uses
sophisticated Natural Language Processing (NLP) techniques, including zero-shot
classification, keyword analysis, and encoded content detection (e.g. base64,
hexadecimal, URL encoding), to detect, decode, and classify harmful inputs; and
(2) A summarization module that processes and summarizes adversarial research
literature to give the LLM context-aware defense knowledge. This approach
strengthens LLMs' resistance to adversarial exploitation by fusing text
extraction, summarization, and harmful prompt analysis. According to
experimental results, this integrated technique has a 98.71% success rate in
identifying harmful patterns, manipulative language structures, and encoded
prompts. By employing a modest amount of adversarial research literature as
context, the methodology also allows the model to react correctly to harmful
inputs with a larger percentage of jailbreak resistance and refusal rate. While
maintaining the quality of LLM responses, the framework dramatically increases
LLM's resistance to hostile misuse, demonstrating its efficacy as a quick and
easy substitute for time-consuming, retraining-based defenses.
| [
{
"version": "v1",
"created": "Fri, 2 May 2025 14:42:26 GMT"
},
{
"version": "v2",
"created": "Mon, 5 May 2025 14:46:48 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Muhaimin",
"Sheikh Samit",
""
],
[
"Mastorakis",
"Spyridon",
""
]
] |
2505.01357 | Zihan Wang | Xinghao Qiao, Zihan Wang, Qiwei Yao, Bo Zhang | Weight-calibrated estimation for factor models of high-dimensional time
series | This version includes the supplementary material of the paper | null | null | null | stat.ME math.ST stat.TH | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The factor modeling for high-dimensional time series is powerful in
discovering latent common components for dimension reduction and information
extraction. Most available estimation methods can be divided into two
categories: the covariance-based under asymptotically-identifiable assumption
and the autocovariance-based with white idiosyncratic noise. This paper follows
the autocovariance-based framework and develops a novel weight-calibrated
method to improve the estimation performance. It adopts a linear projection to
tackle high-dimensionality, and employs a reduced-rank autoregression
formulation. The asymptotic theory of the proposed method is established,
relaxing the assumption on white noise. Additionally, we make the first attempt
in the literature by providing a systematic theoretical comparison among the
covariance-based, the standard autocovariance-based, and our proposed
weight-calibrated autocovariance-based methods in the presence of factors with
different strengths. Extensive simulations are conducted to showcase the
superior finite-sample performance of our proposed method, as well as to
validate the newly established theory. The superiority of our proposal is
further illustrated through the analysis of one financial and one macroeconomic
data sets.
| [
{
"version": "v1",
"created": "Fri, 2 May 2025 15:52:42 GMT"
},
{
"version": "v2",
"created": "Mon, 5 May 2025 03:36:27 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Qiao",
"Xinghao",
""
],
[
"Wang",
"Zihan",
""
],
[
"Yao",
"Qiwei",
""
],
[
"Zhang",
"Bo",
""
]
] |
2505.01370 | Joschka Roffe | Cl\'ement Poirson, Joschka Roffe, Robert I. Booth | Engineering CSS surgery: compiling any CNOT in any code | 43 pages | null | null | null | quant-ph | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We introduce a framework for implementing logic in CSS-style quantum error
correction codes, building on the surgery methods of Cowtan and Burton [CB24].
Our approach offers a systematic methodology for designing and analysing
surgery protocols. At the physical level, we introduce the concept of subcodes,
which encapsulate all the necessary data for performing surgery. At the logical
level, leveraging homological algebra, subcodes enable us to track the logical
operations induced by any surgery protocol, regardless of the choice of logical
operator basis. In particular, we make no assumptions on the structure of the
logical operators of the code, thereby eschewing the irreducibility assumption
that has been necessary in other formulations of surgery for CSS codes [Coh+22;
Cro+24; ZL24]. As a proof of concept, we develop a surgery protocol inspired by
lattice surgery that implements a logical CNOT gate between any two logical
qubits. Applicable to any CSS code, this protocol is highly versatile and
facilitates the process of fault-tolerant design.
| [
{
"version": "v1",
"created": "Fri, 2 May 2025 16:17:47 GMT"
},
{
"version": "v2",
"created": "Mon, 5 May 2025 13:16:49 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Poirson",
"Clément",
""
],
[
"Roffe",
"Joschka",
""
],
[
"Booth",
"Robert I.",
""
]
] |
2505.01410 | Ananth Narayanan | Sebastian Brandt, Ananth Narayanan | Towards Optimal Deterministic LOCAL Algorithms on Trees | To appear at PODC 2025 | null | null | null | cs.DC | http://creativecommons.org/licenses/by/4.0/ | While obtaining optimal algorithms for the most important problems in the
LOCAL model has been one of the central goals in the area of distributed
algorithms since its infancy, tight complexity bounds are elusive for many
problems even when considering \emph{deterministic} complexities on
\emph{trees}. We take a step towards remedying this issue by providing a way to
relate the complexity of a problem $\Pi$ on trees to its truly local
complexity, which is the (asymptotically) smallest function $f$ such that $\Pi$
can be solved in $O(f(\Delta)+\log^*n)$ rounds. More specifically, we develop a
transformation that takes an algorithm $\mathcal A$ for $\Pi$ with a runtime of
$O(f(\Delta)+\log^*n)$ rounds as input and transforms it into an
$O(f(g(n))+\log^* n)$-round algorithm $\mathcal{A}'$ on trees, where $g$ is the
function that satisfies $g(n)^{f(g(n))}=n$. If $f$ is the truly local
complexity of $\Pi$ (i.e., if $\mathcal{A}$ is asymptotically optimal), then
$\mathcal{A}'$ is an asymptotically optimal algorithm on trees, conditioned on
a natural assumption on the nature of the worst-case instances of $\Pi$. Our
transformation works for any member of a wide class of problems, including the
most important symmetry-breaking problems. As an example of our transformation
we obtain the first strongly sublogarithmic algorithm for
$(\text{edge-degree+1})$-edge coloring (and therefore also $(2\Delta-1)$-edge
coloring) on trees, exhibiting a runtime of $O(\log^{12/13} n)$ rounds. This
breaks through the $\Omega(\log n/\log\log n)$-barrier that is a fundamental
lower bound for other symmetry-breaking problems such as maximal independent
set or maximal matching (that already holds on trees), and proves a separation
between these problems and the aforementioned edge coloring problems on trees.
We extend a subset of our results to graphs of bounded arboricity.
| [
{
"version": "v1",
"created": "Fri, 2 May 2025 17:39:13 GMT"
},
{
"version": "v2",
"created": "Mon, 5 May 2025 17:09:18 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Brandt",
"Sebastian",
""
],
[
"Narayanan",
"Ananth",
""
]
] |
2505.01427 | Maksym Shamrai | Maksym Shamrai | Perturbation Analysis of Singular Values in Concatenated Matrices | 11 pages | null | null | null | cs.LG stat.ML | http://creativecommons.org/licenses/by/4.0/ | Concatenating matrices is a common technique for uncovering shared structures
in data through singular value decomposition (SVD) and low-rank approximations.
However, a fundamental question arises: how does the singular value spectrum of
the concatenated matrix relate to the spectra of its individual components? In
this work, we develop a perturbation framework that extends classical results
such as Weyl's inequality to concatenated matrices. We establish analytical
bounds that quantify the stability of singular values under small perturbations
in the submatrices. Our results show that if the matrices being concatenated
are close in norm, the dominant singular values of the concatenated matrix
remain stable, enabling controlled trade-offs between accuracy and compression.
These insights provide a theoretical foundation for improved matrix clustering
and compression strategies, with applications in numerical linear algebra,
signal processing, and data-driven modeling.
| [
{
"version": "v1",
"created": "Tue, 11 Mar 2025 09:28:57 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Shamrai",
"Maksym",
""
]
] |
2505.01428 | Han Yang | Han Yang, Chuanguang Yang, Qiuli Wang, Zhulin An, Weilun Feng, Libo
Huang, Yongjun Xu | Multi-party Collaborative Attention Control for Image Customization | null | null | null | null | cs.CV | http://creativecommons.org/licenses/by/4.0/ | The rapid advancement of diffusion models has increased the need for
customized image generation. However, current customization methods face
several limitations: 1) typically accept either image or text conditions alone;
2) customization in complex visual scenarios often leads to subject leakage or
confusion; 3) image-conditioned outputs tend to suffer from inconsistent
backgrounds; and 4) high computational costs. To address these issues, this
paper introduces Multi-party Collaborative Attention Control (MCA-Ctrl), a
tuning-free method that enables high-quality image customization using both
text and complex visual conditions. Specifically, MCA-Ctrl leverages two key
operations within the self-attention layer to coordinate multiple parallel
diffusion processes and guide the target image generation. This approach allows
MCA-Ctrl to capture the content and appearance of specific subjects while
maintaining semantic consistency with the conditional input. Additionally, to
mitigate subject leakage and confusion issues common in complex visual
scenarios, we introduce a Subject Localization Module that extracts precise
subject and editable image layers based on user instructions. Extensive
quantitative and human evaluation experiments show that MCA-Ctrl outperforms
existing methods in zero-shot image customization, effectively resolving the
mentioned issues.
| [
{
"version": "v1",
"created": "Wed, 2 Apr 2025 12:45:49 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Yang",
"Han",
""
],
[
"Yang",
"Chuanguang",
""
],
[
"Wang",
"Qiuli",
""
],
[
"An",
"Zhulin",
""
],
[
"Feng",
"Weilun",
""
],
[
"Huang",
"Libo",
""
],
[
"Xu",
"Yongjun",
""
]
] |
2505.01429 | Md. Rakibul Islam | Md. Zahid Hossain, Md. Rakibul Islam and Most. Sharmin Sultana Samu | Explainable AI-Driven Detection of Human Monkeypox Using Deep Learning
and Vision Transformers: A Comprehensive Analysis | null | null | null | null | cs.CV | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Since mpox can spread from person to person, it is a zoonotic viral illness
that poses a significant public health concern. It is difficult to make an
early clinical diagnosis because of how closely its symptoms match those of
measles and chickenpox. Medical imaging combined with deep learning (DL)
techniques has shown promise in improving disease detection by analyzing
affected skin areas. Our study explore the feasibility to train deep learning
and vision transformer-based models from scratch with publicly available skin
lesion image dataset. Our experimental results show dataset limitation as a
major drawback to build better classifier models trained from scratch. We used
transfer learning with the help of pre-trained models to get a better
classifier. The MobileNet-v2 outperformed other state of the art pre-trained
models with 93.15% accuracy and 93.09% weighted average F1 score. ViT B16 and
ResNet-50 also achieved satisfactory performance compared to already available
studies with accuracy 92.12% and 86.21% respectively. To further validate the
performance of the models, we applied explainable AI techniques.
| [
{
"version": "v1",
"created": "Thu, 3 Apr 2025 19:45:22 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Hossain",
"Md. Zahid",
""
],
[
"Islam",
"Md. Rakibul",
""
],
[
"Samu",
"Most. Sharmin Sultana",
""
]
] |
2505.01430 | Praneeth Medepalli | Muna Numan Said, Aarib Zaidi, Rabia Usman, Sonia Okon, Praneeth
Medepalli, Kevin Zhu, Vasu Sharma, Sean O'Brien | Deconstructing Bias: A Multifaceted Framework for Diagnosing Cultural
and Compositional Inequities in Text-to-Image Generative Models | Published at ICLR 2025 Workshop SynthData | null | null | null | cs.CV | http://creativecommons.org/licenses/by/4.0/ | The transformative potential of text-to-image (T2I) models hinges on their
ability to synthesize culturally diverse, photorealistic images from textual
prompts. However, these models often perpetuate cultural biases embedded within
their training data, leading to systemic misrepresentations. This paper
benchmarks the Component Inclusion Score (CIS), a metric designed to evaluate
the fidelity of image generation across cultural contexts. Through extensive
analysis involving 2,400 images, we quantify biases in terms of compositional
fragility and contextual misalignment, revealing significant performance gaps
between Western and non-Western cultural prompts. Our findings underscore the
impact of data imbalance, attention entropy, and embedding superposition on
model fairness. By benchmarking models like Stable Diffusion with CIS, we
provide insights into architectural and data-centric interventions for
enhancing cultural inclusivity in AI-generated imagery. This work advances the
field by offering a comprehensive tool for diagnosing and mitigating biases in
T2I generation, advocating for more equitable AI systems.
| [
{
"version": "v1",
"created": "Sat, 5 Apr 2025 06:17:43 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Said",
"Muna Numan",
""
],
[
"Zaidi",
"Aarib",
""
],
[
"Usman",
"Rabia",
""
],
[
"Okon",
"Sonia",
""
],
[
"Medepalli",
"Praneeth",
""
],
[
"Zhu",
"Kevin",
""
],
[
"Sharma",
"Vasu",
""
],
[
"O'Brien",
"Sean",
""
]
] |
2505.01431 | Wenqi Guo | Wenqi Guo, Shan Du | ZS-VCOS: Zero-Shot Outperforms Supervised Video Camouflaged Object
Segmentation | null | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Camouflaged object segmentation presents unique challenges compared to
traditional segmentation tasks, primarily due to the high similarity in
patterns and colors between camouflaged objects and their backgrounds.
Effective solutions to this problem have significant implications in critical
areas such as pest control, defect detection, and lesion segmentation in
medical imaging. Prior research has predominantly emphasized supervised or
unsupervised pre-training methods, leaving zero-shot approaches significantly
underdeveloped. Existing zero-shot techniques commonly utilize the Segment
Anything Model (SAM) in automatic mode or rely on vision-language models to
generate cues for segmentation; however, their performances remain
unsatisfactory, likely due to the similarity of the camouflaged object and the
background. Optical flow, commonly utilized for detecting moving objects, has
demonstrated effectiveness even with camouflaged entities. Our method
integrates optical flow, a vision-language model, and SAM 2 into a sequential
pipeline. Evaluated on the MoCA-Mask dataset, our approach achieves outstanding
performance improvements, significantly outperforming existing zero-shot
methods by raising the F-measure ($F_\beta^w$) from 0.296 to 0.628. Remarkably,
our approach also surpasses supervised methods, increasing the F-measure from
0.476 to 0.628. Additionally, evaluation on the MoCA-Filter dataset
demonstrates an increase in the success rate from 0.628 to 0.697 when compared
with FlowSAM, a supervised transfer method. A thorough ablation study further
validates the individual contributions of each component. More details can be
found on https://github.com/weathon/vcos.
| [
{
"version": "v1",
"created": "Thu, 10 Apr 2025 06:24:54 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Guo",
"Wenqi",
""
],
[
"Du",
"Shan",
""
]
] |
2505.01432 | Chi Zhang | Chi Zhang | Dynamic Asset Pricing: Integrating FinBERT-Based Sentiment
Quantification with the Fama--French Five-Factor Model | 22 pages,2 figures | null | null | null | cs.CE | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | This paper presents a comprehensive study on the integration of text-derived,
time-varying sentiment factors into traditional multi-factor asset pricing
models. Leveraging FinBERT, a domain-specific deep learning language model, we
construct a dynamic sentiment index and its volatility from large-scale
financial news and social media data covering 2020 to 2022. By embedding these
sentiment measures into the Fama French five-factor regression, we rigorously
examine whether sentiment significantly explains variations in daily stock
returns and how its impact evolves across different market volatility regimes.
Empirical results demonstrate that sentiment has a consistently positive impact
on returns during normal periods, while its effect is amplified or even
reversed under extreme market conditions. Rolling regressions reveal the
time-varying nature of sentiment sensitivity, and an event study around the
June 15, 2022 Federal Reserve 75 basis point rate hike shows that a
sentiment-augmented five-factor model better explains abnormal returns relative
to the baseline model. Our findings support the incorporation of
high-frequency, NLP-derived sentiment into classical asset pricing frameworks
and suggest implications for investors and regulators.
| [
{
"version": "v1",
"created": "Tue, 22 Apr 2025 18:45:03 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Zhang",
"Chi",
""
]
] |
2505.01433 | Cong Qi | Cong Qi, Hanzhang Fang, Siqi jiang, Tianxing Hu, and Wei Zhi | Enhancing TCR-Peptide Interaction Prediction with Pretrained Language
Models and Molecular Representations | null | null | null | null | q-bio.QM cs.CL cs.LG | http://creativecommons.org/licenses/by/4.0/ | Understanding the binding specificity between T-cell receptors (TCRs) and
peptide-major histocompatibility complexes (pMHCs) is central to immunotherapy
and vaccine development. However, current predictive models struggle with
generalization, especially in data-scarce settings and when faced with novel
epitopes. We present LANTERN (Large lAnguage model-powered TCR-Enhanced
Recognition Network), a deep learning framework that combines large-scale
protein language models with chemical representations of peptides. By encoding
TCR \b{eta}-chain sequences using ESM-1b and transforming peptide sequences
into SMILES strings processed by MolFormer, LANTERN captures rich biological
and chemical features critical for TCR-peptide recognition. Through extensive
benchmarking against existing models such as ChemBERTa, TITAN, and NetTCR,
LANTERN demonstrates superior performance, particularly in zero-shot and
few-shot learning scenarios. Our model also benefits from a robust negative
sampling strategy and shows significant clustering improvements via embedding
analysis. These results highlight the potential of LANTERN to advance TCR-pMHC
binding prediction and support the development of personalized immunotherapies.
| [
{
"version": "v1",
"created": "Tue, 22 Apr 2025 20:22:34 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Qi",
"Cong",
""
],
[
"Fang",
"Hanzhang",
""
],
[
"jiang",
"Siqi",
""
],
[
"Hu",
"Tianxing",
""
],
[
"Zhi",
"Wei",
""
]
] |
2505.01434 | Kyriakos Giagiakos | Kyriakos Giagiakos | Supervisory Control of a Flexible Manufacturing Unit for the Production
of Two Products | MSc thesis | null | null | null | math.OC cs.FL | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | In this diploma thesis, the mathematical model of a multi-product
manufacturing unit will be presented. The unit consists of a set of three
conveyors, a robot, a lathe, a milling machine, an assembly machine, and a
painting machine. Finally, the connection of the above elements is carried out
via one slot buffer. The products can be divided into two distinct sets
according to the route they will follow through the machines of the unit. The
mathematical models of the individual subsystems of the plant using finite
deterministic automata will be presented. The two different routes that the
products can follow will be presented. The desired product flow will be
presented in the form of desired regular languages. The properties of the
desired languages with respect the overall automaton of the system will be
investigated. A supervisory architecture will be designed based on the desired
regular languages.
| [
{
"version": "v1",
"created": "Tue, 22 Apr 2025 22:04:28 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Giagiakos",
"Kyriakos",
""
]
] |
2505.01435 | Carlo Siebenschuh | Carlo Siebenschuh, Kyle Hippe, Ozan Gokdemir, Alexander Brace, Arham
Khan, Khalid Hossain, Yadu Babuji, Nicholas Chia, Venkatram Vishwanath, Rick
Stevens, Arvind Ramanathan, Ian Foster, Robert Underwood | AdaParse: An Adaptive Parallel PDF Parsing and Resource Scaling Engine | This paper has been accepted at the The Eighth Annual Conference on
Machine Learning and Systems (MLSys 2025) | null | null | null | cs.IR cs.CL cs.DC cs.LG | http://creativecommons.org/licenses/by/4.0/ | Language models for scientific tasks are trained on text from scientific
publications, most distributed as PDFs that require parsing. PDF parsing
approaches range from inexpensive heuristics (for simple documents) to
computationally intensive ML-driven systems (for complex or degraded ones). The
choice of the "best" parser for a particular document depends on its
computational cost and the accuracy of its output. To address these issues, we
introduce an Adaptive Parallel PDF Parsing and Resource Scaling Engine
(AdaParse), a data-driven strategy for assigning an appropriate parser to each
document. We enlist scientists to select preferred parser outputs and
incorporate this information through direct preference optimization (DPO) into
AdaParse, thereby aligning its selection process with human judgment. AdaParse
then incorporates hardware requirements and predicted accuracy of each parser
to orchestrate computational resources efficiently for large-scale parsing
campaigns. We demonstrate that AdaParse, when compared to state-of-the-art
parsers, improves throughput by $17\times$ while still achieving comparable
accuracy (0.2 percent better) on a benchmark set of 1000 scientific documents.
AdaParse's combination of high accuracy and parallel scalability makes it
feasible to parse large-scale scientific document corpora to support the
development of high-quality, trillion-token-scale text datasets. The
implementation is available at https://github.com/7shoe/AdaParse/
| [
{
"version": "v1",
"created": "Wed, 23 Apr 2025 18:38:41 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Siebenschuh",
"Carlo",
""
],
[
"Hippe",
"Kyle",
""
],
[
"Gokdemir",
"Ozan",
""
],
[
"Brace",
"Alexander",
""
],
[
"Khan",
"Arham",
""
],
[
"Hossain",
"Khalid",
""
],
[
"Babuji",
"Yadu",
""
],
[
"Chia",
"Nicholas",
""
],
[
"Vishwanath",
"Venkatram",
""
],
[
"Stevens",
"Rick",
""
],
[
"Ramanathan",
"Arvind",
""
],
[
"Foster",
"Ian",
""
],
[
"Underwood",
"Robert",
""
]
] |
2505.01436 | Qi Duan | Qi Duan, Ehab Al-Shaer | Firewall Regulatory Networks for Autonomous Cyber Defense | null | null | null | null | cs.CR cs.SY eess.SY | http://creativecommons.org/publicdomain/zero/1.0/ | In this paper, we present the principles of designing new self-organising and
autonomous management protocol to govern the dynamics of bio-inspired
decentralized firewall architecture based on Biological Regularity Networks.
The new architecture called Firewall Regulatory Networks (FRN) exhibits the
following features (1) automatic rule policy configuration with provable
utility-risk appetite guarantee, (2) resilient response for changing risks or
new service requirements, and (3) globally optimized access control policy
reconciliation. We present the FRN protocol and formalize the constraints to
synthesize the undetermined components in the protocol to produce interactions
that can achieve these objectives. We illustrate the feasibility of the FRN
architecture in multiple case studies.
| [
{
"version": "v1",
"created": "Thu, 24 Apr 2025 00:43:10 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Duan",
"Qi",
""
],
[
"Al-Shaer",
"Ehab",
""
]
] |
2505.01437 | Hassan Wasswa | Hassan Wasswa, Timothy Lynar and Hussein Abbass | Enhancing IoT-Botnet Detection using Variational Auto-encoder and
Cost-Sensitive Learning: A Deep Learning Approach for Imbalanced Datasets | null | null | null | null | cs.LG cs.AI | http://creativecommons.org/licenses/by/4.0/ | The Internet of Things (IoT) technology has rapidly gained popularity with
applications widespread across a variety of industries. However, IoT devices
have been recently serving as a porous layer for many malicious attacks to both
personal and enterprise information systems with the most famous attacks being
botnet-related attacks. The work in this study leveraged Variational
Auto-encoder (VAE) and cost-sensitive learning to develop lightweight, yet
effective, models for IoT-botnet detection. The aim is to enhance the detection
of minority class attack traffic instances which are often missed by machine
learning models. The proposed approach is evaluated on a multi-class problem
setting for the detection of traffic categories on highly imbalanced datasets.
The performance of two deep learning models including the standard feed forward
deep neural network (DNN), and Bidirectional-LSTM (BLSTM) was evaluated and
both recorded commendable results in terms of accuracy, precision, recall and
F1-score for all traffic classes.
| [
{
"version": "v1",
"created": "Sat, 26 Apr 2025 02:04:30 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Wasswa",
"Hassan",
""
],
[
"Lynar",
"Timothy",
""
],
[
"Abbass",
"Hussein",
""
]
] |
2505.01438 | Tengfei Xing | Tengfei Xing, Xiaodan Ren, Jie Li | Global Stress Generation and Spatiotemporal Super-Resolution
Physics-Informed Operator under Dynamic Loading for Two-Phase Random
Materials | null | null | null | null | cs.LG cond-mat.mtrl-sci cs.AI | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Material stress analysis is a critical aspect of material design and
performance optimization. Under dynamic loading, the global stress evolution in
materials exhibits complex spatiotemporal characteristics, especially in
two-phase random materials (TRMs). Such kind of material failure is often
associated with stress concentration, and the phase boundaries are key
locations where stress concentration occurs. In practical engineering
applications, the spatiotemporal resolution of acquired microstructural data
and its dynamic stress evolution is often limited. This poses challenges for
deep learning methods in generating high-resolution spatiotemporal stress
fields, particularly for accurately capturing stress concentration regions. In
this study, we propose a framework for global stress generation and
spatiotemporal super-resolution in TRMs under dynamic loading. First, we
introduce a diffusion model-based approach, named as Spatiotemporal Stress
Diffusion (STS-diffusion), for generating global spatiotemporal stress data.
This framework incorporates Space-Time U-Net (STU-net), and we systematically
investigate the impact of different attention positions on model accuracy.
Next, we develop a physics-informed network for spatiotemporal
super-resolution, termed as Spatiotemporal Super-Resolution Physics-Informed
Operator (ST-SRPINN). The proposed ST-SRPINN is an unsupervised learning
method. The influence of data-driven and physics-informed loss function weights
on model accuracy is explored in detail. Benefiting from physics-based
constraints, ST-SRPINN requires only low-resolution stress field data during
training and can upscale the spatiotemporal resolution of stress fields to
arbitrary magnifications.
| [
{
"version": "v1",
"created": "Sat, 26 Apr 2025 08:37:29 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Xing",
"Tengfei",
""
],
[
"Ren",
"Xiaodan",
""
],
[
"Li",
"Jie",
""
]
] |
2505.01439 | Surajit Biswas | Surajit Biswas and Bipul Saurabh | Vanishing of dimensions and nonexistence of spectral triples on compact
Vilenkin groups | null | null | null | null | math.OA math.PR math.QA | http://creativecommons.org/licenses/by/4.0/ | We compute the spectral dimension, the dimension of a symmetric random walk,
and the Gelfand-Kirillov dimension for compact Vilenkin groups. As a result, we
show that these dimensions are zero for any compact, totally disconnected,
metrizable topological group. We provide an explicit description of the
$K$-groups for compact Vilenkin groups. We express the generators of the
$K_0$-groups in terms of the corresponding matrix coefficients for two specific
examples: the group of $p$-adic integers and the $p$-adic Heisenberg group.
Finally, we prove the nonexistence of a natural class of spectral triples on
the group of $p$-adic integers.
| [
{
"version": "v1",
"created": "Sat, 26 Apr 2025 13:37:59 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Biswas",
"Surajit",
""
],
[
"Saurabh",
"Bipul",
""
]
] |
2505.01440 | Alkis Sygkounas | Alkis Sygkounas, Ioannis Athanasiadis, Andreas Persson, Michael
Felsberg, Amy Loutfi | Interactive Double Deep Q-network: Integrating Human Interventions and
Evaluative Predictions in Reinforcement Learning of Autonomous Driving | Accepted at IEEE Intelligent Vehicles Symposium (IV) 2025, 8 pages | null | null | null | cs.LG cs.AI cs.RO | http://creativecommons.org/licenses/by/4.0/ | Integrating human expertise with machine learning is crucial for applications
demanding high accuracy and safety, such as autonomous driving. This study
introduces Interactive Double Deep Q-network (iDDQN), a Human-in-the-Loop
(HITL) approach that enhances Reinforcement Learning (RL) by merging human
insights directly into the RL training process, improving model performance.
Our proposed iDDQN method modifies the Q-value update equation to integrate
human and agent actions, establishing a collaborative approach for policy
development. Additionally, we present an offline evaluative framework that
simulates the agent's trajectory as if no human intervention had occurred, to
assess the effectiveness of human interventions. Empirical results in simulated
autonomous driving scenarios demonstrate that iDDQN outperforms established
approaches, including Behavioral Cloning (BC), HG-DAgger, Deep Q-Learning from
Demonstrations (DQfD), and vanilla DRL in leveraging human expertise for
improving performance and adaptability.
| [
{
"version": "v1",
"created": "Mon, 28 Apr 2025 05:25:18 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Sygkounas",
"Alkis",
""
],
[
"Athanasiadis",
"Ioannis",
""
],
[
"Persson",
"Andreas",
""
],
[
"Felsberg",
"Michael",
""
],
[
"Loutfi",
"Amy",
""
]
] |
2505.01441 | Joykirat Singh | Joykirat Singh, Raghav Magazine, Yash Pandya, Akshay Nambi | Agentic Reasoning and Tool Integration for LLMs via Reinforcement
Learning | null | null | null | MSR-TR-042025-V1 | cs.AI | http://creativecommons.org/licenses/by/4.0/ | Large language models (LLMs) have achieved remarkable progress in complex
reasoning tasks, yet they remain fundamentally limited by their reliance on
static internal knowledge and text-only reasoning. Real-world problem solving
often demands dynamic, multi-step reasoning, adaptive decision making, and the
ability to interact with external tools and environments. In this work, we
introduce ARTIST (Agentic Reasoning and Tool Integration in Self-improving
Transformers), a unified framework that tightly couples agentic reasoning,
reinforcement learning, and tool integration for LLMs. ARTIST enables models to
autonomously decide when, how, and which tools to invoke within multi-turn
reasoning chains, leveraging outcome-based RL to learn robust strategies for
tool use and environment interaction without requiring step-level supervision.
Extensive experiments on mathematical reasoning and multi-turn function calling
benchmarks show that ARTIST consistently outperforms state-of-the-art
baselines, with up to 22% absolute improvement over base models and strong
gains on the most challenging tasks. Detailed studies and metric analyses
reveal that agentic RL training leads to deeper reasoning, more effective tool
use, and higher-quality solutions. Our results establish agentic RL with tool
integration as a powerful new frontier for robust, interpretable, and
generalizable problem-solving in LLMs.
| [
{
"version": "v1",
"created": "Mon, 28 Apr 2025 10:42:49 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Singh",
"Joykirat",
""
],
[
"Magazine",
"Raghav",
""
],
[
"Pandya",
"Yash",
""
],
[
"Nambi",
"Akshay",
""
]
] |
2505.01442 | Steffen Schulz | Steffen Schulz | Algorithm Performance Spaces for Strategic Dataset Selection | Bachelor's thesis, 29 pages, 9 figures, 1 table | null | null | null | cs.IR | http://creativecommons.org/licenses/by/4.0/ | The evaluation of new algorithms in recommender systems frequently depends on
publicly available datasets, such as those from MovieLens or Amazon. Some of
these datasets are being disproportionately utilized primarily due to their
historical popularity as baselines rather than their suitability for specific
research contexts. This thesis addresses this issue by introducing the
Algorithm Performance Space, a novel framework designed to differentiate
datasets based on the measured performance of algorithms applied to them. An
experimental study proposes three metrics to quantify and justify dataset
selection to evaluate new algorithms. These metrics also validate assumptions
about datasets, such as the similarity between MovieLens datasets of varying
sizes. By creating an Algorithm Performance Space and using the proposed
metrics, differentiating datasets was made possible, and diverse dataset
selections could be found. While the results demonstrate the framework's
potential, further research proposals and implications are discussed to develop
Algorithm Performance Spaces tailored to diverse use cases.
| [
{
"version": "v1",
"created": "Tue, 29 Apr 2025 12:29:52 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Schulz",
"Steffen",
""
]
] |
2505.01443 | Ilgar Giyas Aliyev | I.G. Aliyev, F.S. Latifov, A.M. Guliyeva | Parametric vibrations of a damaged orthotropic geometrical shell
stiffened with an inhomogeneous rod and rings on viscoelastic medium | 7 pages, 4 figures,International Journal on Technical and Physical
Problems of Engineering | International Journal on Technical and Physical Problems of
Engineering, Iss. 60, Vol. 16, No. 3, Sep. 2024 | null | null | math.OC | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The structural element considered in the presented article consists,
according to the geometric structure, of a coating and reinforcement elements,
according to the mechanical characteristics of heterogeneous coatings along the
length, having damage inside due to their physical structure and, finally, a
system in contact with a viscoelastic medium. Taking into account one of the
existing models (Winkler or Pasternak), the contact conditions between the
coating and the reinforcement elements and the influence of the medium on the
coating, the frequency equation of oscillation was solved, the results were
analyzed. Theory of hereditary type damage under the action of external force
is often used for taking into account the damages in the structure of a
cylindrical shell forced to vibration. The Hamilton-Ostrogradsky variational
principle is used for solving the problem. The results obtained can be used in
the foundations of bridges built across mountain rivers. Such support is
cost-effective. It should be noted that reinforcement and concrete mortar are
also used in the internal parts of the cylindrical supports used. The article
proposes to fill the inside of the cylindrical lid under study with clay.
| [
{
"version": "v1",
"created": "Tue, 29 Apr 2025 14:17:56 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Aliyev",
"I. G.",
""
],
[
"Latifov",
"F. S.",
""
],
[
"Guliyeva",
"A. M.",
""
]
] |
2505.01444 | Alejandro Gonz\'alez Nevado | Alejandro Gonz\'alez Nevado | Order Structures around Evolution Algebras | Thesis submitted, accepted and defended in 2019 to the University of
Malaga for partial fulfillment of the degree of Master in Mathematics under
the supervision of Prof. Dr. Mercedes Siles Molina | null | null | null | math.RA | http://creativecommons.org/publicdomain/zero/1.0/ | We consider evolution algebras and their related substructures: evolution
ideals and evolution subalgebras. After exposing some of the concepts related
to them in the literature, we explore the order structures that arise in the
sets of substructures of an evolution algebra. This leads to the introduction
of the socle of an evolution algebra and to the study of its connection with
some distinguished elements within the algebra (such as idempotents and natural
elements). Finally, we examine the order structures that emerge among these
elements when we consider the substructures that they generate inside the
algebra. Thus, we develop two order-theoretic approaches (one using
distinguished substructures and other using distinguished elements). These two
strategies could be used in many ways. Particularly, we direct them towards the
study of the socle of an evolution algebra.
| [
{
"version": "v1",
"created": "Tue, 29 Apr 2025 14:39:51 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Nevado",
"Alejandro González",
""
]
] |
2505.01446 | Aashish Misraa | Aashish Kumar Misraa, Naman Jain, Saurav Singh Dhakad | Waymo Driverless Car Data Analysis and Driving Modeling using CNN and
LSTM | This work contributed to research acknowledged in
https://doi.org/10.3390/app10062046 | null | 10.3390/app10062046 | null | cs.RO | http://creativecommons.org/licenses/by/4.0/ | Self driving cars has been the biggest innovation in the automotive industry,
but to achieve human level accuracy or near human level accuracy is the biggest
challenge that research scientists are facing today. Unlike humans autonomous
vehicles do not work on instincts rather they make a decision based on the
training data that has been fed to them using machine learning models using
which they can make decisions in different conditions they face in the real
world. With the advancements in machine learning especially deep learning the
self driving car research skyrocketed. In this project we have presented
multiple ways to predict acceleration of the autonomous vehicle using Waymo's
open dataset. Our main approach was to using CNN to mimic human action and LSTM
to treat this as a time series problem.
| [
{
"version": "v1",
"created": "Tue, 29 Apr 2025 19:45:16 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Misraa",
"Aashish Kumar",
""
],
[
"Jain",
"Naman",
""
],
[
"Dhakad",
"Saurav Singh",
""
]
] |
2505.01447 | Zeinab Teimoori | Zeinab Teimoori | LLM-Enabled EV Charging Stations Recommendation | 5 pages, 4 figures, 2 tables | null | null | null | cs.IR cs.ET | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Charging infrastructure is not expanding quickly enough to accommodate the
increasing usage of Electric Vehicles (EVs). For this reason, EV owners
experience extended waiting periods, range anxiety, and overall
dissatisfaction. Challenges, such as fragmented data and the complexity of
integrating factors like location, energy pricing, and user preferences, make
the current recommendation systems ineffective. To overcome these limitations,
we propose RecomBot, which is a Large Language Model (LLM)-powered prompt-based
recommender system that dynamically suggests optimal Charging Stations (CSs)
using real-time heterogeneous data. By leveraging natural language reasoning
and fine-tuning EV-specific datasets, RecomBot enhances personalization,
improves charging efficiency, and adapts to various EV types, offering a
scalable solution for intelligent EV recommendation systems. Through testing
across various prompt engineering scenarios, the results obtained underline the
capability and efficiency of the proposed model.
| [
{
"version": "v1",
"created": "Tue, 29 Apr 2025 19:57:05 GMT"
}
] | 2025-05-06T00:00:00 | [
[
"Teimoori",
"Zeinab",
""
]
] |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.